Interfaces required to multiple domains

I think their decision to keep themselves a software infrastructure company is smart. In this way, they can apply their systems to many market segments where operations are involved. When operations are performed, some kinds of data are generated and often times those data should be collected, stored, and analyzed to tune and improve operations and business processes. In order to dive into new domains, they need to keep adding new interfaces as well as adding and revising in areas they already cover. Dave Roberts told me that they now have close to 500 interfaces.

Coming from the IT segment, I see people tending to converge to a handful of well-defined standards and, therefore, interfaces. When I first put my foot into the data center market, I was very, very surprised to find out that there were many interfaces on the facilities side. Although BACnet is becoming a major force in the data center facilities protocol of choice, there are still several other protocols, such as Modbus and LonWorks, being used. An IT guy like me tends to think we can force facilities to adopt a single standard to consolidate all the protocols into one, which is IP. I now know it does not work that way. I got involved in NIST's Smart Grid Interoperability Panel, which was organized to come up with a set of standards to allow smart grid to function without conflicting technologies and protocols. The power industry has been around longer than IT, and there are many standards by IEEE, IEC, and others. The power industry has been conducting business to keep the lights on for more than 100 years, and they will not listen to IT about consolidating everything to IT technologies and protocols, for sure.

How to translate domain specific requirements for software developers

OSIsoft maintains that their core PI system is generic and does not change when they apply PI to different vertical markets. When they pick a new domain, they add new interfaces specifically required for that domain. So every time they step into a new domain, they need to worry about yet more interfaces to maintain. This seems daunting, but it is the only practical way to have a generic system to apply to many areas, such as the power industry, oil and gas, and building management segments.

For each vertical domain there is a dedicatedindustry management team that includes experts in that field who can communicate natively with customers. The experts get agreements on requirements, then translate those requirements to a specification for software development teams and partners/ecos to work on.

How to enter a conservative industry like the power industry

IT's change of pace is very fast. New technologies come and go quickly, sometimes within months, if not days. In contrast, utility companies are very conservative and do not replace their technologies and equipment for many years until new technologies or equipment are proven to work solidly. I was curious to find out how a software company like OSIsoft could penetrate into the conservative power industry. In the 1990s, OSIsoft partnered with Westinghouse and also with ABB. Through their introductions to utilities, they started to work with utility players. They expanded their market presence in the utilities market. Although there are a lot of similarities, each utility has specific needs, which triggers customization. But OSIsoft does not provide customization services. Customization is done by utilities themselves or system integrators. Nearly all—97%—of their revenue comes from software maintenance; the remaining 3% comes from basic services such as installation. So a highly configurable nature is important for their product.

Sharing data among multiple entities

In general, if two entities work together, it would be most beneficial to share data among the two. For example, let me refer to the power grid in California. California ISO (CAISO), which reliably balances power supply and demand on the transmission, does not maintain the transmission lines. The lines are maintained by PG&E, a local utility in my region that also is responsible for the distribution grid. Power imbalances can be caused by operational or equipment problems. Therefore, it is very useful if CAISO shares data with PG&E so that they can work together to solve the problem. For this, OSIsoft has released a new feature called PI Cloud Connect, which allows highly granular data to be shared with specific accessibility control in a cloud setting. In this way, any number of organizations can share time-series data with a specific access privilege. Yes, this is a good application of ICT.

Analytics

Once data are captured and stored, they are analyzed to derive useful information to improve operations and business processes. Analytics can be done at many levels. They can be as simple as out-of-bounds values analysis all the way up to prediction. Here OSIsoft does not do its own analytics packages but makes sure to plug in others' packages seamlessly to the PI system. I am currently looking into analytics more in detail. Because analytics is a very broad term and it contains so many angles, most presentations or white papers on products do not mention it in detail. That is frustrating, to say the least.

What is an example of analytics in the utilities business?

Analytics example 1: equipment preventive maintenance

Do you see boxes of different colors and shapes on utility poles around you? One of those boxes is called a transformer and is used to step down high voltage to lower voltage before power gets to your home. Most transformers are based on the electromagnetism discipline and degrade physically as time goes by. If a transformer malfunctions or fails, power to your home will be interrupted. It would be nice to know when to repair or replace it before it fails. One of the analytics packages can monitor its health, bounce it with the historical trend, and provide an early warning.

Analytics example 2: wind power generation

Another example is in wind power generation. Wind is hard to predict. It is blowing one moment but not the next. It is vital to balance the demand and supply of power every second. If we cannot predict power generated by wind, it makes it more difficult to balance power. So it is very important to predict when wind blows and when it stops. Predictive analytics is used widely in weather forecasting, and wind prediction is part of it. First, a prediction model is developed from the historical data, and the model is fine-tuned and modified as more data are collected.

Analytics example 3: smart charging for EVs

Currently, in California, power demand increases as the day goes on and hits a peak in the early afternoon. It goes down to its lowest point during the night. An electric vehicle (EV) like the Nissan Leaf or Chevy Volt is known to draw about the same amount of power as a typical household. If they are charged when power demand is at peak, we run out of power to satisfy demand. But during the night, we usually have plenty of power available, and it is suitable to charge EVs at night at home. This is what a typical EV owner does now. As more public charging stations pop up, and faster yet power-hungry new charging technologies proliferate, charging may be done during peak time. That would disturb the power balance and lead to outages. For this reason, smart charging needs to be developed and deployed. The result of this type of analytics would dynamically allow charging to start when supply satisfies demand.

Different utilities could use an analytics package developed by one utility, but OSIsoft does not share particular users' analytics algorithm with others. OSIsoft has its users communities, and those who belong to them might share such an algorithm via community. The T&D User Group community exists for 20 years, and they tend to share information when there is no competition among them.

Analytics example 4: more renewable energy sources for power generation in California

California has adopted a renewables portofolio system, known as RPS. This specifies the minimum percentage of renewable energy sources, like solar and wind, in power generation. California plans to attain 33% of all the power from renewable energy sources by 2020. Although not all the renewable energy sources are highly volatile, like wind power, a lot of unknowns will be thrown into the power grid. Constant power-supply predictions based on ever-changing weather (the wind may or may not blow at any given minute, and solar power goes down when clouds set in) will be vital to keep the power grid stable all the time.

Applying PI to more demanding domains

Smart grid is to make the power grid smarter. Our physical infrastructures consist of more than the just the power grid; we need, for example, gas, water, waste, transportation, government, street lights and traffic systems. Dave is working on the next topic beyond the power grid, which is the smart city. According to Dave, a smart city is defined differently by different people. But currently, US cities like Austin, Seattle, New York, and Chicago have their smart city projects. OSIsoft is involved in some of them, and a public announcement is coming shortly.

Collecting, aggregating, storing, and linking all sorts of data from its different sources would provide tremendous intelligence to a city. A utility at the conference reported that they collect 100,000 data per second. If we implement a system for a smart city, the number of data points would explode by the order of 2 to 3 magnitudes. That means millions of data per second would bombard the PI system. Even though the PI system is created to cope with a large amount of data of many kinds, at some point, they may have to alter their architecture and technologies to process such a massive amount of data. That makes me interested in talking to their technology visionary. Stay tuned for that in a coming blog.

When smart grid was first introduced, Cisco declared that the power grid would be much bigger than the Internet. From the data point of view alone, the amount of data produced and processed on the power grid is on a scale that none of us has experienced before. And with more-sophisticated monitoring technologies, the volume of data will even increase. The data collected may include equipment health, power flow, and quantity of power consumption. Simply collecting data does not do much good. We need to process what we collect—make heads and tails of it—to produce useful information for better operation and maintenance. This is the Big Data problem that is getting a lot of attention these days in ICT and other segments.

Usually, Big Data problems are due to the proliferation of SNSs, such as Facebook, Twitter, and LinkedIn. But with the advent of low-power and low-priced, yet very sophisticated, end devices and sensors, different kinds of Big Data problems are emerging, such as the one I just mentioned.

There are several companies that apply their software systems and tools to solve Big Data problems in a particular vertical market, such as the power industry. When I was covering data centers and their energy efficiency, I visited OSIsoft at its San Leandro, CA, headquarters in 2009. They collect data sent by end devices like sensors and their equivalents and store, analyze, and visualize the collected data to take appropriate actions for improving operations. Since that visit, my focus has expanded to include the power industry, which is only one of the markets OSIsoft addresses (see the other markets here).

Recently, I had an opportunity to attend their users conference in San Francisco.

I listened to several representatives of utilities and others in the power industry talk about their use of OSIsoft's PI system. I also talked to Dave Roberts, Fellow and market Principal – Smart Cities, who is an expert in the power industry.

Dave Roberts

The following is my summary of our discussion, with my comments.

Some power grid basics

I am targeting this blog to very, very IT people and not to power people. So I think very simple, basic information is useful. The power grid is a big connected network of power lines. The power grid consists of two types of grids: transmission and distribution. Generated power is transmitted at a very high voltage via transmission lines to neighborhoods of consumers. Then the high voltage is transformed to much lower voltage, and power is delivered to consumers like you and me via the distribution grid. Because power must be consumed as it is produced, demand and supply need to be balanced all the time. Power on transmission lines is managed by each utility or by organizations called ISOs/RTOs (independent of utility companies) to make sure the balance of demand and supply is maintained—to keep the lights on. Also, as with computer networks, it is important to know the health and status of each device and all the equipment hanging from the grid. As in computer networks, such information is collected from multiple places in the grid. The number of collection points grows as more technologies are developed.

What OSIsoft does

Architecture

Although from my conversations with other OSIsoft people, I knew what business they were in, I just wanted to make sure who they are and what they do. They provide a software infrastructure system called PI to connect remote devices, gather/collect/aggregate data from them, and store and retrieve the collected data for further analysis, such as data analytics and visualization. They do not provide end devices like sensors or analytics engines. In other words, PI is one of the important components of the Internet of Things, M2M, or intelligent systems. Different people define the Internet of Things, M2M, and intelligent systems slightly differently, and the terms are often used interchangeably.

Here's an oversimplified view of PI architecture.

My view on the conceptual view of PI architecture

PI is not an operating system but there is some analogy between PI and Windows. Windows provides a base operating environment for applications to run in. Microsoft in general does not provide any applications packages but provides this base plus some tools/utilities and libraries via APIs. Third parties exploit this platform to write applications. PI is similar and does not provide applications, including data analytics packages. So PI can be said to be a general platform and applications area agnostic.

This will continue to Part 2.

An ecosystem for intelligent systems consists of several players, such as chip, OS, middleware, end device, cloud service, back office processing and analytics providers, and system integrators. Ayla Networks, which is still in stealth mode, claims that they provide secure connectivity for an end-to-end solution for an intelligent system. They currently focus on the consumer market but do not rule out expansion into other areas.

I sat down with David Friedman, CEO of Ayla Networks, during the recent Design West to find out what they are up to.

David Friedman

Who they are

David was VP business development for a wireless chip company before. After selling it in 2010 , he saw a business opportunity. At that time, end devices were beginning to be connected to form the Internet of Things. But the ugly reality was that those thousands of end devices were very different from each other, with microcontrollers in a variety of architectures and operating systems, as compared with the nonembedded world dominated by Windows and Linux. All those differences sure were a hindrance to accelerating and proliferating the Internet of Things. David and his cofounders saw the need for a generic solution that could absorb these differences. That led to the formation of Ayla Networks.

David and his team started to work on his solutions. Using his background as a chip guy, he teamed up with STMicro because ST is a major player in the microcontroller market. Ayla Networks is a software company and does not deal with hardware, so this is a good combination. Chip vendors focus on how to design and develop new and better chips but are not experts in networking such as Berkeley sockets and SSL. In other words, companies should focus on their core competency and outsource the rest. In the same vein, application vendors are not experts in the lower layers of software that support applications. During Design West, I heard from several players that application vendors should outsource the lower layers and concentrate on their core business; that is, design and develop applications.

So David is saying "Come to us. We will absorb any protocols differences and security needs to support your applications. You do not need to worry about the lower layers and other infrastructure concerns.”

What they do

Ayla provides an end-to-end connectivity software; for example, to remotely control your AC from outside your home with your smartphone. If you implement something like that on your own, you need to develop lower-layer software for the smartphone, including secure interfaces with its OS and networking stack. Then you need to develop an application to work with that infrastructure. Then you need to worry about how to connect it to your target AC. Communication can be via cellular, WAN, LAN, or PAN. You need to choose the right one. Finally, on the target AC, some mechanism needs to be incorporated to receive data and control from your smartphone. For that, a small board with a communications chip on it must be inserted along with the lower-layer software. And as with your smartphone, you need to interface with that chip's OS and networking stack, on top of developing applications.

What Ayla provides:

1. Client-side lower-layer software for applications
2. Networking solutions with security
3. Cloud services
4. Lower-layer software on target appliances

The client-side software can be integrated with your applications and downloaded from Apple Store and Google Play like other applications. Ayla provides whatever networking protocols are required by the applications. In addition, they provide cloud services to connect your client devices to target appliances. David did not elaborate on how they provide such services. Cloud services consist of cloud infrastructures and applications in the form of virtual machines. Because of the proliferation of inexpensive cloud infrastructures services, a startup like Ayla can afford to provide the cloud services. The lower-layer software on target appliances is the same as #1. Application developers can focus on their core business of developing applications without getting bogged down in lower-layer stuff.

Team

Now this seems to require a lot of technical expertise in several areas, such as embedded systems, networking, and cloud. Although these areas are closely related, no one person could address all of them. Although David did not reveal details about his team, he did say that he gathered technical people who had worked together for several years. People who like to innovate and have a passion to create something new are attracted to his team.

Devil is in the details

Many people have discussed controlling an AC from outside with a smartphone or a tablet, and that by itself is nothing new. David told me that now is the perfect time to bring their solutions to the market. Technologies have advanced and the market is opening up. An article by Reuters reports that by 2022 a typical household will own 50 Internet-connected devices, compared with 10 now. David said that we do not want 50 solutions for 50 devices but only a single solution so that any new device can easily belong to the existing network. He also emphasized that creating a supportable product is really, really difficult.

Their infrastructure pieces must be easy to:

• install
• configure with a lot of latitude
• scale
• implement with secure delivery

They claim that they have met all four requirements.

Big Data

They are in a perfect position to collect and aggregate data, but David did not reveal any future plan for business exploiting such a position. But he did not rule out the possibility, either. If I were an AC OEM, I would be very interested in analyzing control data sent by smartphones, to reflect on how to tune my AC features. David told me that the key to the use of Big Data is anonymization with the ability to opt in or out.

Energy consumption

What about power consumption? Smartphones eat a lot of power, and additional features like these would consume even more. David told me that his developers pay a lot of attention to curbing power consumption. Power-use optimization implemented as in the iPhone would attain energy efficiency.

We chatted a bit about power in general when everything is connected. My view was as follows:

1. Advantages: There are many advantages to deriving useful information from generated data that might be otherwise discarded. Some information can be used to save power.
2. Disadvantages: Unless we can intelligently select which data to collect, or keep, or discard, we will end up with a pile of useless data occupying a lot of storage and server equipment, wasting energy.

I think what David said about the disadvantages was interesting. He said that analyzing a vast amount of data, transforming them to a small number of useful data, and discarding the rest might do the trick. I do not know the doability of such a thing, but it is an interesting thought.

Future

David did not give me any concrete future plans, but this system can expand beyond the consumer segment to the commercial and industrial markets. I think there is a reasonable level of traction

in the consumer market at this point, and there will be greater demand later. In addition to a clear application for turning an AC on and off, I can think of a few more examples. Sprinklers for lawns are usually on timers, and occasionally they start to work even in the rain while you are not at home. Your remote device can override this. Or better yet, you can program sprinklers in conjunction with moisture-detecting sensors buried in the ground and with sensors for other local weather.

But I think the really big applications are in the commercial and industrial segments. I think it is very smart of Ayla to choose the consumer market first. There are two reasons. The first is that the commercial and industrial segments are known to be late adopters. The second is that if you target very specialized and sophisticated industry-grade equipment, how many people will know? But familiar appliances like ACs show up on many people’s radar screens; after success in the consumer market, Ayla can enter bigger markets.

The conversation stayed at a high level because they are still in a stealth mode, but a public announcement is forthcoming. Meanwhile, you can register to purchase their design kit by visiting here.

I may cover that session later but offer a few takeaways now:

• Many people who were surveyed expect the market for intelligent systems to grow, but they are not using them yet.

• Application vendors would like to concentrate on applications rather than the lower support layers. In other words, concentrate on your core value and outsource the rest.

So I wanted to talk to someone who does the lower layers, a.k.a. infrastructure. I was lucky enough to talk to Wind River.

Although I never used their VxWorks before, I knew the company before it became a part of the Intel family. They were a specialist in embedded systems with the VxWorks operating system and platforms, including middleware. Then they expanded their product lines to include their versions of Linux and Android. They are now providing infrastructures for intelligent systems for end devices, although they do cover other areas, like gateways and networking, as well.

End devices come in many sizes and functions, as shown in the figure above, such as smart vending machines and digital signage. Smart vending machines can collect data about which products sell well, and signal when stock gets low. The general trend is that end devices and equipment are getting smarter or more intelligent, which is the driving force behind intelligent systems. Wind River is addressing the need for those end devices to be more intelligent.

To prepare for a meeting with Wind River, I read several articles by the company, published here, and listened to a Wind River presentation.

              Presentation made by Wind River (Source: Wind River)

The following is a summary of my conversation with Wind River. I hope it helps my readers understand intelligent systems, still a very young market.

What are the differences among M2M, intelligent systems, and the Internet of Things?

I started to hear about M2M a few years ago, and a little after that the Internet of Things became widely known. Finally, intelligent systems appeared on my radar screen. I just want to note that there is no entry for intelligent systems in Wikipedia. The current entry refers to a company with that name. In that entry, there is a link saying, "For the computer science phenomenon, see artificial intelligence.” But I do not think that the intelligent systems discussed here are the same as artificial intelligence. This may not mean a lot, but it indicates that the term is so new that there is no independent entry for it in Wikipedia yet.

Are they different? Based on my quick search and Wind River's definition, M2M refers to technologies to connect end devices via networks, whether wired or wireless (PAN, LAN, or WAN). But the difference or the similarity of the Internet of Things and intelligent systems is not clear. Actually, those three things are being used loosely in the marketplace, as this field is so new and is still evolving. My own take is as follows. M2M focuses on device connectivity. Connecting end devices became possible, and that in itself was a big deal. And then it led to a new phenomenon called the Internet of Things. Then people realized that connecting devices made the entire system more intelligent because the injection of intelligence became possible, hence the term intelligent systems. As we have a more connected society, whether the connection is human to human, human to machine, or machine to machine, more intelligence will be injected at many places.

So it seems that connectivity led to the formation of intelligence. Then what is intelligence?

What is intelligence?

Does connectivity alone generate intelligence?

Take my favorite example, a smart meter. The power meter you and I have at home used not to be very smart. It simply measured our power consumption and recorded it. A human meter reader showed up once a month to read it to find out how much we consumed, regardless of when we consumed it. A smart meter collects power consumption information every hour and sends it to your regional utility. I challenged Wind River by saying that adding connectivity to a meter alone does not generate intelligence. A meter with connectivity does not seem any smarter than a traditional meter. Both measure power consumption. I have a tendency to nitpick some details that many people take for granted. Their answer was very good and really articulated what intelligent systems are all about. They said connectivity makes it possible to inject intelligence because it is not possible to do so without connectivity.

Let's take an example they used. By adding connectivity to a meter, it is now possible to implement new things, including the following:

1. Time-of-use pricing, which can distinguish when you use power. The price during the night is the cheapest, and the price at peak time (usually early afternoon) is the highest.

2. Smarter (optimized) operations of the power grid that may eliminate the need to build more power plants.

3. A service delivery platform to the home.

A smart meter alone could not realize these functions. But its connectivity injects a new set of functions into the entire power grid, and that is the injection of intelligence that makes the power grid system more intelligent and brings it closer to being an intelligent system.

Other intelligence

Is there any other intelligence? In addition to the above, data collected from smart meters and other parts of the power grid will be analyzed with analytics packages and useful information may be derived. That information can be used to improve the operations and maintenance of the power grid. That is intelligence. But Wind River is not currently addressing this type of intelligence.

What's wrong with the roll-your-own approach?

As I said before, unlike the IT market, each vertical market needs a specific infrastructure, including many different kinds of networking protocols. Wind River recommends that you not implement your own solution but use the infrastructure of someone whose job it is to develop and maintain such infrastructures. Those in a specific vertical market should concentrate on their applications (core value) rather than infrastructures (nonessential value). This is the same message I heard from one of the vendors during the early intelligent systems market analysis session. This sounds very reasonable, and I think their message is a good one.

But I have two reservations. One is specificity. If you are in a vertical market and require a specific infrastructure but no one can provide it, what can you do? Like OSIsoft, Wind River monitors the market and adds necessary infrastructure support as needed. No single company could support each and every interface. What reasonably can be done is to select major ones and support them. The next figure shows a partial list of standards for different markets supported by Wind River.

                A partial list of standards for different markets supported

                                by Wind River (Source Wind River)

My other reservation is that Wind River and the Intel family are not the only company that provides such a solution. What if, in the future, you need to interact with organizations that use other vendors' solutions? There is no guarantee that the systems will be interoperable.

I know it is not a fair question when the market is still in its infancy. From my past involvement with the Smart Grid Interoperability Panel (SGIP) by NIST, which defines standards for smart grid technology interoperability, I know that standardization efforts will be required. Towards that end, Intel has announced its Intelligent Systems Framework (ISF). I hope that over time, through input from others, such a framework will grow up to be a standard that allows any end device and equipment to be freely interoperable in intelligent systems.

Energy efficiency by intelligent systems

Do intelligent systems contribute to energy efficiency or energy conservation?

Wind River stated here:

Ecological Considerations: Machines can perform power management tasks with finer precision and faster response times than manual, human-dependent systems – saving energy, prioritizing usage, setting policies for response to outages, and the like.

That says it all.

Can we be optimistic?

I was somewhat pessimistic about the settlement for interoperability in the near future, as I have experienced SGIP's work and see many existing standards across different vertical markets. Wind River was bullish for such a settlement. I knew it was not a fair question but asked Wind River about a timeframe. They said they were not certain but thought that in five to ten years some kind of movement would occur for interoperability. I certainly hope they are right. I, for one, want to have such interoperability to exploit real intelligent systems that not only make our society convenient but also promote energy efficiency.

It is necessary to understand the infrastructure of a data center before you can manage it effectively. There are a few classifications given in the area of how data center infrastructure is managed. One informal categorization might be inventory, change, capacity, simulation, and efficiency modeling, although some analysts use more comprehensive categories.

One basic aspect of data center infrastructure management (DCIM) is to measure and monitor what's happening in a data center. To run a data center effectively, we need to know what's in the data center (asset management) and how each component is functioning, including its status and consumption of energy (measuring and monitoring). It sounds trivial to add a sensor to each component and poll its condition regularly. We can interrogate and manage each component as well as the aggregate to grasp the entire status of a data center from a dashboard. Describing it at a high level is straightforward and simple, but the devil is in the details. There are several vendors in this space, and I have talked to some of them, like SynapSense, Sentilla, and Aperture, in the past.

Recently, I had a chance to speak to Richard Jenkins, VP Marketing of RF Code. RF Code manufactures RF tags and sensors as well as the software to process the data they collect, and markets them as an integrated system. Their solution tracks assets, and monitors the environment around the assets in a data center.

Asset management is one of the basic functions of a data center. Not confined to a data center, corporate assets need to be tracked from time to time because they may be lost physically or be hard to locate without close tracking. In the past, asset management was done manually on a spreadsheet. When there are many pieces to track in a data center, manual tracking requires an enormous effort and is not practical. On top of that, equipment, especially in IT, gets moved and replaced constantly. Without some automated means, it is next to impossible to track each piece's location correctly.

Also, it is essential to monitor each piece of equipment and measure data relevant to its operation, because each element in the data center must function flawlessly to ensure reliable operation. Without automated means, it is almost impossible to do so for the large number of components residing in a data center.

It is relatively easy to claim that you have a solution in asset management and monitoring at a data center by deploying sensors and the software to manage them. So I asked Richard about RF Code’s differentiation. Every company claims that its solutions and products are unique and stand out from the competition. His answer was twofold.

Product layering: One part is a common infrastructure to track and monitor elements in a data center. In a single infrastructure, different tags and sensors hang together, ranging from sensors for humidity and temperature to those for motion. This is well described in the following figure.

In this common hardware infrastructure, they put in a higher layer of software for more sophisticated processing.

Open API: Expanding this common product infrastructure philosophy, I would like to classify two major trends for integration. The DCIM market is growing but still rather confusing. There are many aspects to managing a data center. A few analysts have defined a model and areas of coverage. In addition, an increasing number of vendors have rebranded their tools and utilities as solutions for DCIM. Because no single vendor could provide a complete and comprehensive solution for the entire data center operation, large and small vendors alike have started to partner together to provide comprehensive solutions.

Some may provide nonstandard APIs only common to their partners to integrate their tools and utilities to work seamlessly together. The advantage of that scheme is that you have good coverage of tightly integrated DCIM functions as long as you make that particular group as vendors of your choice. Also, if that group's APIs become standard, it would be great. The downside of that is that you may become locked in to that group of vendors. They may not have some functions that may become necessary for you later. If other ways of integrating functions and APIs become standard, you may need to adjust your solutions accordingly.

Another way is make your solutions interoperable with established standards, such as network/serial data formats, and web-based standards like XML and JSON. Those well-established standards tend to be at a higher level, and your level of integration may be looser than in the former approach. However, the upside of this approach is to guarantee that your solution would be integrated with any other tools or utilities that conform to the established standards and would be future proof.

Both approaches are valid because we cannot tell how the DCIM market will shape up in the future. RF Code selected the latter approach.

Wireless communications: The second point Richard raised was wireless communications. When you see racks of IT gear in their surroundings, you probably see many cables for networking and power alike all over the place. If we place sensors in all the strategic locations, you would be inundated with many more cables for communications and sourcing for power. Managing those cables alone would add a burden for data center operators. RF Code, as its name indicates, manufactures and markets only wireless sensors with no cables. A battery-powered wireless node sounds like a great idea.

But it creates its own problems in general:

1. Scalability with wireless communications crosstalk

2. Security

3. Battery life

4. A large effort in tagging

As for #1, their radio complies with the ISO 18000–7: Air interface at 433.92 MHz. When you deploy a large number of nodes for communications, interference tends to happen and accurate communications may not be guaranteed. RF Code has developed a set of technologies to pack wireless nodes in a way that prevents interference among them and has obtained seven patents in this area to increase the number of nodes without interference. Their patents are listed here. The details are beyond the scope of this blog.

As for #2, some security-sensitive organizations, such as financial institutions, do not want to use wireless communications because of the potential security risk. Unlike in wired communications, packets can be easily grabbed in wireless communications and the information in them stolen. Wireless communications can be protected via encryption (by something like SSL), but encryption eats bandwidth. RF Code has an option not to encrypt the communication but to send very proprietary data. It would still be possible for someone to make a reader that could read their tags’ beacons if hackers really wanted to "sniff” the data. However, aside from the tag-identifier data (basically, a serial number) and current sensor reading (for sensors, or for asset tags that feature tamper detection/motion detection/IR receivers), there is no other information about the assets included in the beacon. So even if an asset tag is affixed to a very important piece of equipment, its beacon data doesn't look any different from data affixed to anything else—it's just a number. All tag-ID-to-asset correlation is done in the backend stem, which would clearly be secured behind a firewall. RF Code’s customers include a number of banks, such as Lloyds Bank in the UK.

As for #3, each of RF Code's wireless nodes is powered by an internal battery with a life of somewhere around five to seven years. That means that every five to seven years a battery must be replaced. In a data center, a server's life is about that long or even shorter. When a battery requires replacement, the server is also replaced. That is why I think this may not be a big issue.

Also note that some of their asset tags equipment and all of their sensors are user serviceable, enabling the user to replace the battery. They also include a "low-battery warning” feature that will alert the admin that the battery is getting low when it gets down to about 20% charged.

As for #4, as the amount of equipment grows, the effort to tag it all grows as well. It would be a major effort to tag an existing data center. When equipment has been deployed already, reaching each piece may be cumbersome and difficult. IT equipment like servers may be stored inside a rack, and extra effort may be required to attach a tag at the right location. Some servers may have an outlet exhaust on the side rather than on the back. Without proper placement, a tag may be influenced by the sideways exhaust. Also, without detailed documentation, tagging information may be only on the equipment's label and nameplate, which may not be very precise or adequate. For example, it may not be easy to find out which department particular equipment belongs to and what it is for.

In any event, it would be much easier to place a tag on each piece of equipment and device before its deployment. RF Code gets involved in an early stage of data center construction and avoids this problem. But if people need to tag assets that are already in place, they recommend that they do so as part of a typical annual inventorying process that requires staff to physically account for each individual asset. Since they have to do this for accounting purposes anyway, tagging assets as part of that process presents the least additional effort.

Expanding the idea of tagging as early as possible before the data center is in operation, let's consider a container-based data center. A container-based solution has several advantages. Those include ready-made modular additions to data center capacity and less garbage as a result of not having to unwrap each component. When a container-based solution is put together, a tag can be applied to each piece of equipment as it is assembled into the container. At the time of the assembly, it is easy to reach any location for tagging. Also, the information on each piece of equipment is readily available, and each tag can contain precise information.

RF Code works with IBM and HP to integrate the data they collect with their software system. Although large companies like IBM and HP have many divisions and each division sometimes behaves like an independent company, both companies have a container-based data center solution. Incorporating RF Code’s solutions into their container-based data centers would improve their deployment.

Finally, I asked Richard two questions: about the DCIM market and their future plans.

DCIM market present and future

Present:

Richard and I were in agreement that the DCIM market is poorly defined and is very confusing. He felt that DCIM started to receive recognition only in the past 12 months. He also mentioned that DCIM is a poorly formed acronym. He thought it was more like data center management infrastructure (DCMI) than DCIM, which typically tracks assets and monitors power consumption and environmental conditions. It also needs to integrate with building management systems. But that is not enough. On top of what DCIM provides, DCMI needs to have functions for software loads, networking, and hardware management. With those additional functions, DCMI could provide real-time proactive management of a data center.

I agree with his idea. Many of the currently available solutions mostly touch the facilities side and have very little impact on IT equipment. Even if they touch the IT side, it is only to look at each piece of IT equipment as a black box and not to deal with what's happening inside, such as loading factor, software status, virtualization, and software execution efficiency. The reason for the omission is simply the difficulty of measuring and monitoring such data and analyzing and incorporating it into the dashboard for visualization and good integration of IT and facilities. I am glad that a vendor like RF Code has a good vision of DCIM similar to mine.

Future:

As for the near future of the DCIM market, I asked Richard if he thought some companies will dominate the market. It’s crowded right now. Many vendors market their solutions as DCIM, and they come in many sizes and functions, such as established companies like Intel and Schneider. Smaller companies like RF Code provide niche, pure DCIM solutions. Richard thought that for the time being, the DCIM market will be dominated by a combination of large established companies and startups. I agree with him. Using the business intelligence market as an example, he predicted how the market might shape up. Smaller companies will be merged or acquired by larger ones, and consolidation will take place as it did in the BI market. At some point, DCIM will be one of the functions in the larger infrastructure management solution market.

RF Code’s future directions

RF Code applies their technologies to the data center segment as well as to the oil and gas market. I thought their technologies could be applied to the power industry. The power grid consists of a large assortment of devices and equipment. Smart grid is an attempt to merge power, communications, and IT technologies into a cohesive system to increase the effectiveness of power generation, transmission, distribution, and consumption. Along the power grid, there are many assets deployed to support each function, and it is important to track them accurately. For example, after a power outage is confirmed, it is necessary to locate which device or equipment is at fault and identify its location so that a service crew can be sent to repair it.

The power grid must be maintained to guarantee reliable operations to keep the lights on. Each device and piece of equipment needs to be monitored for its proper operation in real time. As each becomes electronics-based from the electromagnetic base, each component will be controlled in a more precise manner, and tracking its location and status will be more important. I think RF Code’s products could be adjusted to be applied to the power industry, but each vertical has its own idiosyncrasies in vocabulary and operations. It may not be so easy to do. They do not have a plan to expand into this market at this time.

They instead would like to grow into the higher stack to increase functions with software.

Final thoughts

Although people have started to realize the importance of managing the data center infrastructure in the US, it is not clear whether, or what kind of, comprehensive solutions should be employed. Because the market is still young and no clear standards are set, people are hesitant to invest in a comprehensive solution. But they want to deploy a piecemeal solution that brings a visible result. The DCIM market outside of the US, like APAC, is still being formed. When I talked to a large data center provider, it was not clear to them what DCIM is or what DCIM covers. DCIM is an important market and will grow, but more education on its merits will promote it further.

What does it mean?

How does the IT scene change with the implementation of hybrid cloud computing? First let's consider private clouds only. In the following, I will use an enterprise data center and its private cloud interchangeably for the ease of discussion, although not all data centers have been converted to private clouds yet. Some company may have several data centers (and therefore private clouds) in the US, or even worldwide, across multiple time zones. So even before talking about hybrid, using this technology we can combine those physical data centers into one single logical private cloud. A logical cloud consists of physical private clouds (data centers) and may be recognized as one entity.

Logical private cloud

With a logical private cloud, using some technologies from CloudVelocity, we can move applications that may consist of physical machines (PMs: not virtualized) and virtual machines (VMs) anywhere and anytime we choose. In the following figure, we can pass PMs and VMs back and forth seamlessly between our home cloud and any other private clouds of our company. Although it shows only a subset of interactions below, we can potentially move PMs and VMs in any way that makes sense by some predetermined criteria. It may that one PM or VM is passed to another cloud and then to the third one and so on. It would become pretty complex to manage your Pms and VMs under such a new paradigm.

PMs and VMs move around only among private clouds owned by the same organization. A set of such private clouds may be considered as one logical private cloud.

This means we can finally implement several things discussed in part 1, including:

• Follow the sun

• Follow the moon

• Load-balancing

• Fail-over/disaster recovery

Follow the sun

In a given workday, access to software applications and utilities running on servers and other IT equipment—and therefore clouds—fluctuates. Access starts to grow as people start their day’s activities in the morning, it hits a peak, then subsides towards the evening. Access is lowest during the night. So you might want to move your PMs and VMs to other time zones where the sun still shines and more loads need to be processed. We can expect a better response time when loads and processing units are close to each other.

Follow the moon

In many countries, power is cheaper during off-hours (normally nights, hence follow the moon). Sending your Pms and VMs to such time zones may reduce your operation cost. Additionally, even within the US, power cost can fluctuate hourly if a variable power pricing model is implemented and applied to data centers. By shifting your VMs to a data center whose region gets the lowest power cost, you may save on running costs.

Load-balancing

Just as we load-balance among servers at a data center, we may want to send loads to several different private clouds. In this way, when one data center gets very busy, such loads can be passed to other data centers to share the burden. How you move PMs and VMs should be determined by predefined metrics to optimize your operations for a few factors, such as operating cost, response time, and throughputs. Each organization has its own goal for its operation, and the metrics should be tailored to accommodate it.

Cloud bursting may be related to load-balancing, although it is not the same. When a load increases in a private cloud, we may want to move all or part of it to a public cloud for on-demand processing; this is known as cloud bursting. PMs and VMs that are processing the load can be moved to a public cloud for continuous processing. When the load subsides, PMs and VMs on the public cloud can be disabled. There has been a lot of talk about cloud bursting, but now it can become a reality. We need a good automated system to move PMs and VMs, and to enable and disable them as needed. A good policy is a must-have for this as well.

Fail-over/disaster recovery

The San Francisco Bay Area will have a major earthquake some day for sure, and when it happens, much of the existing infrastructure, including data centers, will be unusable. If we have a way of duplicating what we are running in our primary data centers at a secondary site far enough away (such as the Sacramento area, a little more than 80 miles from the Bay Area) and transferring execution state information intact to the distant site, processing could proceed without interruption.

Super logical private cloud

With this technology, we do not have to consider the boundary between private and public clouds either. So the logical private cloud can include public clouds, becoming a super logical private cloud, or what I call a supercloud.

A green oval depicts a private cloud, and a light-blue one represents a public cloud.

This configuration would make managing PMs, VMs and clouds much more complex. We can move our PMs and VMs between private clouds, between private and public clouds, and among public clouds. We will no longer be restricted to a move between one cloud and another cloud (a one-to-one move) but can implement one-to-many and many-to-many as well. Then it will become necessary to develop a system that allows automation. As we involve many private and public clouds of various implementations, we will not be able to easily track how to optimize such moves. For that, we will probably need a policy based on predefined metrics. Cost may be the number one factor. But at the same time, we want to maximize response time—and the performance of developers scattered around the globe.

Also, note that many superclouds may share the same private and public clouds. This means that loads at each private and public cloud could fluctuate over time. So depending upon how busy each cloud is, we may want to dynamically alter how we form a super logical private cloud for optimization.

By the way, when a supercloud is developed and deployed, will we call it a supercloud or simply a cloud? Those IT folks who will follow us in the future may take it for granted and consider it a normal IT deployment and execution environment. Throughout IT history, when some technology or method becomes transparent as part of an overall system, that is when we say that that technology really has matured.

Who uses hybrid clouds and benefits from them?

I can think of three parties, although there may be more.

Enterprise end-users

Enterprises that have their own private clouds can extend them to public clouds to produce hybrid clouds to exploit the things I mentioned above.

Data center providers

If you are a colo provider, you can sell extra services at your center to realize hybrid clouds for your clients. There are different levels of providers. Some may simply rent a space, while others provide both equipment and services. Some may provide both private and public clouds at the same data center. For them, this is a perfect tool to increase their revenue.

Third-party consultants/service companies

If a colo provider does not want to provide any service other than space, those guys with the hybrid cloud technology can help end-users implement hybrid clouds.

Energy consideration

Finally, my blog always ends with a question about what the subject means to energy efficiency. Although inconclusive, there has been some discussion about whether cloud computing is more energy efficient than its predecessors. I think it depends upon whose view you take. If you are a user, you pass some or all of your computing needs, along with support staff, software, hardware, power, cooling, water, and other things, to your cloud provider on an on-demand basis. Since you can reduce your investment on these, it is certainly energy efficient for you. It may or may not be for your provider. If the provider has very little utilization of their facilities, they may not be profitable or energy efficient at all. You may still have to have a large staff, a large space, dedicated IT and facilities equipment, facilities support such as cooling, and so on. That cannot be very energy efficient.

When a hybrid cloud becomes a supercloud and our energy becomes more scarce, we may need to look at energy consumption and energy efficiency at the supercloud level without distinguishing private or public clouds, which may sound silly at this point. It is because the US seems to be doing fine for the foreseeable future with shale gas and oil, but who knows what may happen next?

Some technical details

As a former technologist, I wanted to open the hood and find out more about the underlying technologies. For this, Anand Iyengar, CloudVelocity’s founder and CTO, gave me a chalk talk.

Anand Iyengar

Because this is not a white paper detailing the technology, I only describe it at my layman’s level. However, it is such an intriguing technology that I’m accepting Anand’s offer for further discussion and will write more about it in the future.

Anand elaborated on the details, but I made a simpler diagram to fit the space. It is not that much different from the picture above.

Virtual machines (VMs) move between a typical enterprise private cloud (mostly VMware-based) and a public cloud (typically Amazon AWS). (Source: CloudVelocity)

Setup:

Let’s take a quick look at the architecture:

Private cloud

• We first look at your own data center or colocation facility (private cloud). In the modern software application system, an application does not run on a single server. Instead, the running of an application spans multiple physical and virtual machines. So we call it a multisystem application. The configuration may differ according to usage and design. Typically, it consists of load balancers, web servers, application servers, and sometimes a cluster of other servers.

• This is illustrated in the figure above. To save space, I drew only two machines, S1 and S2. The multisystem application typically uses a database, file systems mounted from a closed-box NFS server system (NFS1), and services from an LDAP server (LDAP). Everything in the public cloud is a copy of what is in the private cloud, including NFS1. Note that NFS provides files locally but not over the cloud boundary. Moreover, in the private cloud there is a server, such as an LDAP, that one may not want copied to the public cloud but kept in the private cloud for security reasons.

• There are virtual appliances (CloudVelocity Nexus Site Manager for the private cloud and CloudVelocity Cloud Manager for the public cloud) that together keep the cloud site images synchronized with the most recent changes to systems in the private cloud. CloudVelocity uses the term appliance to emphasize its dedicated function. CloudVelocity Nexus may run on a physical server, while CloudVelocity Cloud Manager runs as a virtual machine.

• Let's further assume that S1 (in the VMDK file format) is virtualized, but neither S2 nor DB1 is virtualized.

Public cloud

• Everything in the public cloud is a mirror image of what is in the private cloud. The public cloud is populated by copying what is in the private cloud. An initial copy is made for each system, and updates are sent afterwards.

A. System S1, which is virtualized needs to be copied to a pubic cloud. S1 is copied via the link to the public cloud, unless there is a copy left over from a previous need, in which case only the differences are copied. It is converted to AMI automatically. In the case of S2, it must be copied via the link to the public cloud. Like S1, if there is not a copy left over from a previous need, it gets virtualized to run on an AMI file format.

B. System DB1 and NFS1, which are physical servers, go through the same process. They also are automatically virtualized to run on AWS/AMI.

• The two clouds are linked by the Internet or a dedicated connection via SSL.

• When any of the systems are no longer necessary, they can be disabled and deleted, or retained for future use. The copy may be retained to minimize copying time in the future.

Some high-level description continues regarding how those components work together. The actual workings are much more complex, but I have simplified them for this presentation.

Operation

• CloudVelocity Nexus inventories all the pertinent information regarding computing power in the private cloud, including applications and supporting servers, such as file systems and databases. The configuration information is stored in a proprietary file format.

• Inventory information is passed to the CloudVelocity Cloud Manager in the public cloud. This appliance is virtualized to run on AWS (in the AMI file format) all the time. Storage and computing time for this appliance are charged per AWS pricing. The size of the appliance is negligible at several hundred kilobytes, and it does not cost much. Once Cloud Manager receives the configuration information, storage volumes for each component get allocated for each system and populated, without running it. This reduces activation time for the public cloud counterparts. EC2 charges are heavier for computing than for storage. The design is a good compromise for reducing copying time and saving on computing charges on EC2.

• Starting the systems in the public cloud typically takes three to five minutes, which is the time required to boot up a VM in the AWS cloud. They are started in parallel.

• The systems may be disabled when not needed in the public cloud. The user may expect another need for the systems sometime soon and keep a copy around, or delete it to save the storage charge by the AWS system.

Application areas:

1. Cloud fail-over: If the private cloud goes down for any reason but the operation cannot be halted, a full, earlier copy of the application systems may be started in the public cloud to take over the operation. This is called cloud fail-over and can be used for disaster recovery and for implementing features like follow-the-sun and follow-the-moon.

2. Development and testing sandboxes: More than one full copy of the application can be started simultaneously in the public cloud, while the application is still running in the private cloud. These copies are fully sandboxed and can be used for development or testing.

3. Complete move: For datacenter space constraints and other reasons, the systems in the private cloud may be cloned to the public cloud and those in the private cloud, disabled.

4. Cloudbursting: This allows extending computing power in the private cloud by enabling and cloning computing power in the public cloud, if a load surge takes place. This can be accomplished without losing data integrity in the private cloud, because two appliances can tunnel update requests back to the local site. Any changes made on the public cloud are constantly sent back to the private cloud for data consolidation, so when the load surge subsides and the copies in the public cloud are taken down, data integrity is maintained.

Patent-pending technology

Anand said that two technologies in One Hybrid Cloud Platform (OHCP) are unique, and CloudVelocity is applying for a patent for each.

The first has to do with synchronizing two data stores via two appliances that contain the inventory of computing equipment in both clouds. I will not go into detail, but according to Anand, replicating and maintaining synchronization between the two requires some work. During switchover time between the primary and the secondary copy of a VM by vMotion, pages dirtied on the primary copy are constantly sent to the secondary copy for synchronization. This requires fast (about 5 to 10 ms) communication between the primary and the secondary, but it allows a game running on one server to run continuously on another server after the move. The OHCP sends all the changes once in the form of a file and that makes it possible to send over a slower connection like the Internet with encryption (SSL). As for the moving of a running game, OHCP does not support such a feature.

The second is concerned with letting the duplicated copies of VMs in the public cloud have access over the connection to databases like LDAP in the private cloud. As noted before, because of security concerns, some servers and databases may not be duplicated in the public cloud. So VMs in the public cloud need to have access to them in the private cloud.

OHCP vs. vMotion

After discussion with Anand, I came to understand that vMotion and OHCP address different problems, but may overlap in some functionality. Both technologies move systems in execution from one cloud to another. But there is more to it. I summarized the differences in the following table.

Looking at the table above, it appears that the two technologies are not competing but can be complementary to each other. I will dig into them more in my future blogs.

By the way, I can try out their system free of charge. But wait! I am not ready. I do not have a reasonable-size private cloud myself, much less use AWS. I probably need to consult with some of my friends who are involved in Silicon Valley Cloud Center.

(Continued to part 3, which will discuss energy efficiency by cloud computing and what it means to have a hybrid cloud.)

So this time I tossed my digital camera and PC out of my bag and experimented with my iPad. With the iPad, I do not need a mouse or USB memory. Its charger is smaller and much lighter. Here’s what I found out.

Pros:

1. The shoulder bag got much lighter (1.8 pounds vs. 4.2 pounds). This is helpful. If you are in the US, you can carry a heavy bag from your car for the short distance to your place of meeting. In Japan, you walk a lot, and the heavy bag is not very convenient, especially in crowded trains at rush hours.

2. The always-on feature helps me to use it without booting up. If I don’t use audio or video applications, it lasts long enough to get to Japan (some ten hours).

3. Even without an Internet connection, I can check calendars and documents via DropBox and Documents.

4. Support for English and Japanese works flawlessly.

5. Power consumption of the iPad is less. It consumes 12W while my HP PC uses 65W.

Cons:

1. An iPad is basically a read-only device and is not really suitable for writing, except for short messages in email and other documents.

2. An iPad does not support USB memory unless in a special format, which does not help me very much. I do not want to upload sensitive files to the Internet via DropBox.

Now how did I cope with the two cons? This may not work for you unless you have a situation similar to mine. I travel to Tokyo and Osaka. I have close relatives living in these cities with a PC and an Internet connection. So if I need to use a PC, I borrow theirs and both problems above are solved.

From an energy efficiency point of view, an airplane can be lighter, consuming less fuel, and power consumption by an iPad is lighter, too.

With the two problems fixed, I would be happier. Incidentally, on this trip, I converted one of my relatives from a PC to an iPad. In Japan, Apple iPads are still very popular. She does not carry her PC around or write long documents, and she has minimum exposure to computing technologies. So far she is happy.

Some people project 2013 will be the year of the cloud, and hybrid clouds are talked of as one of the trends for the year to come. See here, here, here, and many other places.

As I said before, much of hybrid cloud is just talk and not reality, and there have been several showstoppers before now.

Some of the many factors making it hard to implement hybrid clouds are mainly technical:

Technical problems

Virtual machine (VM) file format

1. Public cloud: Amazon Web Services was the first to implement a public cloud, and AWS is now the de facto standard for public cloud. It uses its own proprietary file format (Amazon Machine Image, a.k.a AMI) running virtual machines on the Xen hypervisor. Their file format is not the same as the original Xen VM format. So even if you are running Xen hypervisor for your cloud, you cannot enjoy interoperability with AWS without converting your VM's file format. For example, Citrix virtualization environment is based on Xen, but its file format is virtual hard disk (VHD), which is also the file format for Microsoft's virtual machine.

2. Private cloud: In the enterprise market (private cloud), VMware's VM file format (VMDK) is the de facto standard.

3. Hybrid cloud is an attempt to use both private and public clouds to process IT demands by optimizing suitable in-house and outsourced IT infrastructures as needed. So when we want to move VMs back and forth between public and private clouds, we need translations each time we move them across the cloud boundary. It may not be very hard to do so, because there are some translation tools readily available from vendors like Amazon and VMware (vmkfstools). It may be straightforward to move VMs that are not in execution, but VMs in execution are generally hard to move with their execution state intact. See the next.

Physical movement of VMs

• If we want to exploit public and private clouds for an application in execution, that execution instance may be transported between two or more clouds to find the most suitable execution environment. One big issue is the distance between clouds. VMware's vMotion allows you to transport your VM up to something like 100 km (80 miles) but no farther. With this physical restriction, what you can do with hybrid cloud may be limited by the distance between clouds.

Various support environment

• Cloud is not just virtualization but needs a comprehensive environment, such as management and support, including tools and security considerations. Each cloud tends to come with its own environment and idiosyncrasies, so what you can do easily in one cloud may not be as easy in another cloud. This would make managing a hybrid cloud cumbersome.

To date, most discussions on hybrid have been at a very abstract level and not at all concrete. People have talked about what we could do with hybrid cloud without referring to its concrete implementation. Recently, I came across yet another brand-new cloud company that claims to have solved the aforementioned problems. Greg Ness recently sent me email with a press release and wanted to show what CloudVelocity, his new company, is doing in the area of hybrid cloud.

I am by no means an expert in hybrid cloud computing or any kind of cloud computing, for that matter, but let me try to review how hybrid computing is implemented with their technologies. To support hybrid cloud, VMs need to move back and forth between private and public clouds. How can we implement such a move? Because an execution space is not shared between a public and a private cloud, we cannot literally move a VM across the clouds. What we do is to make a copy of a VM executing at one cloud and transport its execution status to a cloned VM at another cloud. Then we can disable the original VM and enable the cloned one. If a VM is not in execution, it is not that hard. But if it is in execution, it is much harder.

If both private and public clouds are implemented with the same technologies and the distance is less than, say, 100 km, the same VM could be transported with a utility like vMotion. But in most cases, two cloud environments are not the same (see the technical problems described above), and the distance could be greater. Also, you can move only virtualized applications but not traditionally maintained applications, because you cannot assume all the applications have been virtualized into a VM format.

We need to have carbon copies of VMs and non-VM versions of applications (that need to be virtualized) on the other side. That means you need to have carbon copies of your applications running on a public cloud. This sounds like a disaster recovery (DR) system.

Disaster recovery/fail-over system

In such a system, you duplicate the applications that are running at the primary location and operate them with options at the secondary location. These options include active-active and active-passive configurations. Active-active means that the machines (and thus applications) are live at both the primary and the secondary locations at the same time, with data being copied from the primary to the secondary sites. In this scenario, when the primary location cannot operate any longer for any reason, the secondary location can take over seamlessly. The active-passive configuration may not guarantee complete synchronization, because the passive one in the secondary location does not run until the primary location can no longer support applications.

In any event, if we duplicate the whole thing for the secondary site, as in the case of DR in an active-active fashion, the duplicated copies are always in the secondary site with dedicated servers. This situation is the farthest from cloud computing in spirit, especially for public clouds.

What we need is a solution like this:

• Copies on the other side made only when needed (on-demand).

• Noninteroperability problems overcome:

• Resolve VM file format and other incompatibilities among major cloud systems, such as AWS, Rackspace, Microsoft, and OpenShift.

• Handle physical vs. virtual applications in an IaaS cloud environment.

Now back to CloudVelocity. I visited Greg Ness and Rajeev Chawla, CEO, at their headquarters in Santa Clara. They claim to have implemented a solution to solve the problems discussed above.

From left: Rajeev Chawla (CEO) and Greg Ness (VP Marketing). See here for their bios.

They have developed a comprehensive system for implementing hybrid cloud that they call One Hybrid Cloud Platform (OHCP), which is depicted in the following picture. Applications move across the cloud boundary in five steps:

1. Host discovery—Inventory your private cloud (data center), which consists of all the pertinent IT hardware and software.

2. Blueprinting—Create a database of how the discovered components are put together.

3. Cloud provisioning—Duplicate and create VMs on the target cloud (translating VMs and virtualizing physical applications if necessary).

4. Synchronization—Synchronize VMs between the two clouds.

5. Service initiation—Let the duplicated VMs take over and disable the original VMs.

            CloudVelocity's comprehensive One Hybrid Cloud Platform.

This sounds easy. How do they do this? That will be covered in Part 2.

Fujitsu hosted the conference, and attendance was free and included breakfast, lunch, and cocktails. Over the years, the number of attendees has grown, and ICT analysts, such as Gartner and IDC, were in the crowd. On top of that, they got very prominent speakers, like Dr. John Hennessy, president of Stanford University, and Nicholas Negroponte, founder of MIT's Media Lab. It is a good deal, to say the least.

The following is my take on what was presented.

Fujitsu develops products over a wide range of areas, including hardware and software, as well as services.

Fujitsu covers many areas, as shown in the picture above. Many Japanese companies are considered good at hardware but not software. But Fujitsu actually does pretty well in the areas of software and services as well as hardware. Because they put their foot in many areas and their base is ICT, it is very interesting to see how they view the current state of ICT. ICT used to stand by itself without much regard to other areas, like energy. In the past, ICT technologies alone could generate revenues. But things have changed a lot recently, and ICT needs to find other application areas. O.K., the following is how Fujitsu sees the world in conjunction with ICT.

This picture shows that ICT could be applied to areas like food/water, economy, energy, population/aging, health, natural disasters, and transportation. Cloud computing would tie them all together. If they are right about this, there are still a lot of application opportunities for ICT to generate revenues, which is good news to many people in the ICT field, including myself.

Following the first keynote on Fujitsu's business, Stanford’s Hennessy gave a talk and there were three presentations by Fujitsu people. I covered Hennessy's presentation in a previous blog. The Fujitsu presentations were categorized as future solutions for smart energy deployments, which is very relevant to what I look at these days. Along with these three technologies, a total of 23 technologies were demoed. First, let me touch on the three presentations.

Energy management system (EMS)

The first was about Fujitsu's energy management system (EMS). In 2011, electricity saving by visualization was implemented to cope with the power shortage after the big quake and the shutdown of the 50 nuclear plants in Japan. From 2012 to 2013, Fujitsu developed a cloud-based building energy management system (BEMS). From 2013 to 2015, it plans to move its focus to smart city. An example of that is to cut power consumption by observing peak times and by controlling battery charging and discharging of laptops at offices. Their experiment showed that they could reduce total office power consumption by 2–3% by doing this. In most building energy management systems, attention is usually given to high-power consumers like HVACs. It is very Japanese to even pay attention to laptop battery charging. But who knows? Small savings may add up to a big saving.

OpenADR

The second was Fujitsu's implementation of OpenADR.

Fujitsu was the first Japanese company to participate in the OpenADR 2.0a interoperability test. Demand and response (DR) is one of the easiest ways—by shaving off power at peak time—to generate logical power. Usually, when demand increases, supply must be increased to cope with the higher demand. Instead, demand is curtailed to fit the supply at a given time. In a way, logical power was generated to solve the demand-and-supply imbalance. OpenADR is a protocol specified by the OpenADR Alliance to dictate in what form the DR signal is transmitted. The member companies are listed here. I chatted with a person who was manning the booth. Their product is a demand response automation server (DRAS) paired with a client. Because OpenADR is a protocol specification, each vendor could build their products based on the standard communications protocol for competition. Fujitsu's involvement is basically in the US; Japanese utilities are not ready to consider DR.

Efficient power supply

The third Fujitsu presentation was about the power supply, which has a conversion rate of 94.8%. An organization called 80 Plus promotes a high conversion rate for the power supply. Even with 80 Plus Platinum, however, the conversion rate is only 90%. Fujitsu developed a few technologies to increase efficiency. Fujitsu plans to release a product with this technology in 2014.

I had to be elsewhere so missed the subsequent sessions. I think ICT has a lot of potential in many areas. Energy is one, and Fujitsu seems to understand it very well.

Long ago, I read one of his papers but this was my first time to see him on the stage. Fujitsu has cultivated working relationships with many prominent research labs and universities throughout the world. Stanford is one of them.

Everyone knows that both Stanford and UC Berkeley incubated many excellent technology companies in Silicon Valley. Many came from Stanford, like SGI, Netscape, Cisco, HP, Yahoo, Google, and Sun, and then there was TCP/IP per Vint Cerf. So Stanford must have done something right.

Dr. John Hennessy

Hennessy’s talk was about transference between universities and the real world. But this could be applied to the enterprise world as well, as in spin-outs and spin-ins. In any event, let me summarize his points, some of which I have paraphrased, and ponder on how they might be applied to the real world.

First, he said new technologies were building blocks for new industries and therefore economic growth. Those technologies came from company and university research labs. He gave examples:

· Semiconductors—initiated by Bell Labs.

· Workstations—remember Xerox Parc’s Alto Computer?

· Internet—no explanation required.

· Web—made possible by the Internet, then made the Internet even bigger.

· Biotechnology—fundamental research was done at universities.

He continued by saying that universities are a source of:

· Discontinuous innovations

· Talent

Two points he wanted to make in his talk were:

1. How to promote innovation in a university or company.

2. How to transfer innovations to real products.

He made the first point clear by saying that it is important to pick the right set of people and let them drive the direction of research. Those people should include visionaries (who have compelling ideas), explorers (who give some direction to those ideas), and uninhibited executors (who carry out the more concrete tasks and often are students.) His example of a visionary was Jim Clark, who founded both Silicon Graphics and Netscape. I had a chance to meet with Jim Clark when he founded Netscape with Marc Andreessen many years ago. Because of some interesting circumstance I will not discuss here, he gave a chalk talk just for me about how he saw the Internet and why a web browser business would grow along with the Internet’s growth by analyzing other players, the telecom and communications market at that time. It was very nicely delivered, and I felt he was a real visionary.

Also, regarding his first point, Hennessy said three things are important:

A) Live on the edge of technology.

B) Find discontinuities.

C) Find the best people.

Although there are subtle differences, I think points A) and B) are very similar in that it is important to consider what could be done even better without getting bogged down by current constraints or the way things have been designed. For A), Hennessy gave as examples both Xerox Alto, which led to the creation of workstations and PCs, and Cisco, which laid the basis for Internet infrastructures. Other examples included the Internet and Yahoo. For the second example, he listed MIPS, which he founded.

Finding good people is also an important factor. At Stanford, more than half the engineering students are from outside the US. Companies like Silicon Graphics, Yahoo, Google, Sun, and VMware were cofounded by people who came from outside this country. Those good people should be given incentives for innovation and freedom to choose their direction.

In the second point of transferring innovation, the biggest point is to transfer people but not technology. This is because:

· Inventors know the technology best.

· New technologies need advocates.

We may have known this intuitively, but he stated it clearly. Come to think of it, it makes good sense. Hennessy thinks students are good ingredients for the transfer. Get good students, grow the talent pool, and teach them how to succeed in entrepreneurship.

He then touched on entrepreneurship. Entrepreneurship alone does not work. It is also important to prepare fertile soil for startups. For this you need:

· Tolerance to smart failure

· Access to capital

· Access to engineering talent

· Access to management talent

· Access to advice

· Manufacturing capability

These requirements have been discussed in many places and in many different ways, but it was good to hear them discussed in this context. He pointed out they even apply to internal transfer.

He concluded his talk by saying that:

He did not mention cleantech, but his idea could be applied to that segment.

When eBay released the information on their new data center (after their Mercury project in Phoenix) in Utah in June, media jumped on it and reported intensively. In addition to the US data center market, I have covered the data center market in Japan, which is seldom reported in the US media. (I set up a session and acted as MC to discuss Japanese data centers in the recent Silicon Valley Leadership Group 5th Data Center Energy Summit. But that is another story.) This is partly because of the Japanese media and data center players in Japan only report their news in Japanese. Unfortunately, the Japanese language is not spoken outside of Japan and the foreign media is hungry for information but has no use with the news reported in Japanese. So I have been encouraging Japan Data Center Council (JDCC), a consortium of data center operators, vendors and users, to make their information available in the US.

Most of major Japanese data centers were not damaged by the earthquake in 2011 but after that, Japan's power supply climate got cloudy as it does not have a solid plan as to what to do with the halted (except 2 of 50) nuclear reactors and hunting for new energy sources. As such, Japanese data center operators are searching for new power sources. I brought fuel cells to their attention as a potential energy source for data centers. Fuel cells that were dismissed as experimental a few years ago are now gaining some good tractions among data center operators, thanks to eBay, Google and Apple for their use at data centers. JDCC wants to find more about the appropriateness of the fuel cells for data centers. I made an arrangement for them to visit eBay to find out and to find out how my friend Jeremy Rodriguez was doing with his new role.

The following is a summary of our conversations with Jeremy and his associate Serena DeVito, and the information that is publicly available.

From left: Jun Sato (Mitsubishi), Dai Tojima (NEC), Serena DeVito, Jeremy Rodriguez and Hideki Okita (Hitachi)

eBay's data center in Utah is the latest one after the one in Phoenix (Mercury project). There are two phases of the construction. The first phase is called Topaz and it has conventional configurations, such as power supplied by local utilities with a 30 MW substation on their premise, UPS boxes and diesel generators. Quicksilver, the second phase, which is attached to Topaz, was to be constructed with the same conventional way. But they decided to try something new. Instead of laying out rows of UPS boxes inside and generators outside, eBay decided to install Bloom Energy Servers. The next two figures indicate two configurations, conventional (UPS boxes and generators) and new (fuel cells) ones

Conventional configuration with UPS boxes (inside) and generators (outside) (courtesy of eBay)

New configuration with fuel cells (courtesy of eBay)

eBay plans to install 30 fuel cells from Bloom energy. A fuel cell unit produces electricity through a chemical reaction fueled by either natural gas or biogas. For the Quicksilver project eBay will be purchasing biogas, however because biogas isn't available to the site they will install pipe infrastructure and consume natural gas from the utility grid.

The Bloom Energy Servers (fuel cells) are not the only greener energy option eBay is pursuing in Utah. When they originally looked at sourcing third party, off-site renewables for the Topaz data center, they found that the law in the State of Utah would not allow it. They worked with a Utah state senator to pass a bill in 2012 that changed this law, and are now in the process of accepting proposals for as much as 20 MW of additional renewable energy. Also, with the fuel cell installation, UPS boxes or generators were no longer necessary and space was released for Bloom Energy Servers that took much smaller space. Smaller footprints and less or no CO2 emissions sound pretty good.

As for the power requirements at the Utah data center, 30 200kW (total of 6MW) Bloom Energy Servers will be installed with the capability to install up to 30MW total in the future. While the power load to Quicksilver ramps up, any excess from the 6MW will be used to power the existing Topaz facility. As eBay plans to expand the data center, it probably requires more than 6MW. Because each Bloom Energy Server is 200kW in power capacity, they can add as many boxes as they need according to varying needs.

So far so good. Advantages seem to be very convincing:

· Cleaner power

· Local generation with less delivery losses

· Smaller footprints

However, I had a six million dollar question. Data center operators are very conservative and do not want to use any unproven technologies unless they have been proven to work by many practitioners for many years. The idea to source power from Bloom Energy Servers with the grid as a backup is a total reverse of the traditional design. On a casual look, this may look reckless but I think they took a calculated risk. I understand the final decision of going with this design was made by CEO. If everyone is risk averse, nothing gets improved. I have met and talked to Dean Nelson who is in charge of data center design/construction/operations at eBay and who can make a bold decision with a solid team of people under him like Jeremy and Serena. eBay, along with the likes of Apple and Google, is doing a great service to the entire data center industry to open up a new frontier. Do you recommend this solution to JDCC member companies? Come back in 18 months and ask the question. That was their answer.

After the meeting at a coffee shop, the members of JDCC and I were sharing the excitement from the meeting. Someone said that the solution is only viable where a lot of open space is available. But even in the urban setting, UPS boxes and a diesel generator occupy some space in the building. If they can be replaced with fuel cell boxes that are smaller in size, this could be a viable solution. Well, I cannot wait 18 months for eBay's answer!

Energy mix for power: Power can be generated from multiple sources, such as coal, oil, gas, and renewable energies (wind, solar, geothermal, and hydro). In the 1970s the SVP began to diversify its energy sources to stabilize its power cost, because the cost of each source is independent and varies. This policy was established when the SVP experienced the high price of oil during the embargo in the 1970's. So Santa Clara’s municipal utility began to have long-term, fixed-price contracts with different energy providers to achieve stable costs for power generation. The next figure denotes its current energy sources.

Energy mix for generation of power for SVP, including hydro, geothermal, coal, solar, natural gas, and wind.

Renewable energies and RPS: The diversity mentioned above contributed to the current requirement for renewable energy use for power generation. In California, by 2020, each power producer should generate 33% of its power from renewable sources. Now 20% of power is generated from renewables and that will rise to 25% in a few years. They are ahead of the pack, and they are providing power generated from renewable sources to others with a renewable energy certificate.

Data center efficiency: In SVP territory, competition drives energy efficiency. To win in the colo market, price is the most important factor. To that end, operators tend to adopt best practices in their data centers to reduce power consumption. The SVP provides rebates for power reduction and provides environmental impact guidelines, such as the LEED program for green buildings.

Impact of cap and trade: The federal version of cap and trade (CT) died in 2009 in the US. In Japan, the Tokyo Metropolitan government began in 2010 to impose its own version of cap and trade, similar to the UK's carbon reduction commitment (CRC). At first, the Tokyo government’s power curtailment requirements did not distinguish between types of business. The Tokyo version is more strict than Japan's national version. While not many other industries are growing, the data center industry is, and it requires more power every year. If the Tokyo cap-and-trade laws were strictly applied, a typical large data center would be required to pay $1M a year in fines. The JDCC felt threatened by this and started a dialogue with the government.

There are many servers and other equipment scattered around in corners of every building in the Tokyo area. Those are managed with yesteryear’s technologies and operational methods, wasting huge amounts of energy. If they were consolidated into data centers with proper operations, more energy could be saved. The JDCC took time (more than six months) to convince the government to apply a different rule to data centers. Even better, now the Tokyo government is promoting the use of data centers rather than a small server room in each company.

Even though the federal version of CT died, the California version of it will be enacted on January 1, 2013. The SVP needs to start preparing for it. So do data center operators in SVP territory and elsewhere in California.

Cost for infrastructure: As more data centers come to Santa Clara and demand more power, new infrastructure must be put in place or existing infrastructure expanded. Who bears the cost for such expansions? In principle, data center owners pay for the substations and other infrastructure. One reason for that is that operators tend to overestimate their needs. If the expansion is done on the basis of their inflated estimates and not fully utilized and the burden is spread among all the ratepayers, it is not fair to other ratepayers. JDCC said that the situation is similar in Japan. In Japan, data center operators must pay for a necessary substation and sometimes for any cables required for connection.

Rate: A data center consumes a lot of power, and the JDCC has a hard time negotiating a favorable purchase cost. The SVP has several levels of rate structure linked to the amount of power consumed: the more you use, the more favorable your rate. Because each situation is unique, the SVP may draw up a special service agreement that is renewed every three years (shortened from 5 years) because of rapidly changing California laws.

Power supply path regulation: In Japan, each data center is given only one power path (feed) from utilities. When a data center is constructed in phase and/or in modular fashion, that is not convenient for them. As they grow, it is favorable to have multiple feeds, but that is not allowed in Japan. If they insist, they can get two separate feeds from two different substations, but it would cost them a bundle. In SVP territory, it is usually one feed for one parcel. So if a data center set aside space (a parcel) for their data center site, they get only one feed. The entry point might be a substation placed at a data center. If the operator likes, he can have multiple feeds from the substation. If a data center operator wants two feeds coming from two distinct substations, that can be honored easily because SVP territory is small and substations are not far from each other.

Power sourcing via transmission line: In Japan (as in the US), some generation facilities are far away from the urban areas where most power is consumed. The Tokyo Electric Power Company (TEPCO), which serves Tokyo and its surrounding areas, has two nuclear power plants far from Tokyo. By the way, those two sites are completely shut down at this time. I remember a power crisis in California back in 2001. My house in PG&E territory experienced blackouts, while my office in Santa Clara did not. The SVP sources power via transmission line. Now-defunct Enron stopped many of the generators under its control and issued a lot of fake transmission-use requests to saturate the transmission line. The SVP was in the same boat but caused only one blackout. This is strictly because it sat down with its large consumers (90% of ratepayers are business/industry) to reduce their power consumption in return for an assurance that they would not be turned off. Because of this, the SVP avoided the next seven or eight blackouts that took place in other parts of California.

Blackouts: For a data center operator, blackouts are a scary thought. Data centers can operate without power supplied by the grid by using backup generators for 24 to 72 hours. But if fuel runs out, what happens? Atsushi Yamanaka of IDC Frontier shared some horror stories from right after March 11, 2011. TEPCO exercised rolling blackouts right after the disaster for a few weeks. Fortunately, there was no blackout in the Tokyo region, but surrounding areas were not so lucky. Data centers where the rolling blackouts were exercised had to run their backup generators, which are mostly gas, rather than diesel-based, turbines and last 24 to 48 hours. Most of the operators had a premium contract with fuel suppliers, but when there was big demand from other many premium account holders, it was not easy to secure enough fuel. The confusion and gridlock in traffic prevented trucks from reaching data centers quickly. On top of that, earlier rolling blackouts did not specify which areas would be affected until the day of their exercise, so it was very hard to plan.

Power provisioning time: In Japan, when a data center operator asks for power to build a new data center, utilities answer is about 24 months. Eventually, that period became 10 months, and sometimes 8 months. It has been said that data center construction time in the US is about 12 to 18 months from beginning to operation. The SVP can provision power in 6 to 9 months, quicker than other utilities.

The JDCC wanted to visit longer, but unfortunately we ran out of time. I sat with them in their meetings during their stay, and this meeting with the SVP was one of the highlights of their trip. I hope the information obtained will be useful to them when they talk with their utilities about securing enough stable power to secure their 24x7 operations. Finally, I would like to thank SVP’s people for their time and information.

In order to propose an alternative pricing structure and better working relationship with utilities, the JDCC wants to find out what other countries are doing about utilities support for data centers. I arranged a meeting with Silicon Valley Power (SVP) to find out. The SVP is not an IOU but a department of the City of Santa Clara. Municipal utilities do not exist in Japan. When I explained what the SVP is and why they would be of interest to the JDCC, I pointed out the following:

• It is a municipal utility.

• Many data centers are within the city limits.

• They accommodate data centers’ unique requirements.

The first point was interesting enough, but the last two really attracted the JDCC’s attention. There should be a reason why so many data centers (around 30) call Santa Clara home. The city is not huge, only 19 square miles. Data center density is pretty high.

Front row, from left: Wendy Stone, Kathleen Hughes, and John Roukema. Back row, from left: Hideki Okita (Hitachi), Atsushi Yamanaka (IDC Frontier), and Jun Sato (Mitsubishi).

Overview: When Silicon Valley had semiconductor factories, those produced a 24x7, flat, stable power demand. Pricing was relatively easy because demand was constant. The load factor was around 74%. Then manufacturing left the Valley, but soon the dot-com era arrived. To support that, in the 1990s data centers began to pop up. Even after the dot-com bust, the trend continued. While demand from other industries goes down, demand from data centers has been rising 2% annually. Demand was at its peak in 2006 but has declined since, thanks to the energy efficiency exercised at data centers.

Attractiveness: There are a few reasons why data centers are coming to Santa Clara. Several criteria determine data center siting, but two must-haves are power and communications networks (fiber). The market may come next, along with other things, like access to a sophisticated workforce and tax considerations. The SVP provides the lowest cost for power in California and has its own dark fiber network, and there are other local fiber providers. So without any visible recruiting efforts, data centers are coming to Santa Clara.

Fiber network laid and maintained by SVP within the city limits of Santa Clara.

Continues to part 2.

One day several years ago I had the urge to go back to Unix/Linux. By then, Red Hat no longer had a cheap solution. I had lost expertise in Linux and confidence in my systems administration skill. I needed an easy Linux solution, which I knew was an oxymoron. I researched a lot about Linux, and tried a few of the many versions of it, and found that confusing. On the basis of reputation, testimonials from others, and my trials, I picked Ubuntu. Hard-to-understand and difficult-to-deal-with were important when I was a techie. But spending several years as a manager changed my attitude towards computing a lot. Computing should support me, but I should not support it. I selected Ubuntu, and my current version is 12.04 LTS (long-term support) version.

When I signed up as press for the recent ARM TechCon, I got a lot of emails from PR people of various companies for interview opportunities. When you have limited time, you review each company's news closely and decide which companies to talk to. It is hard to choose whom to interview with the sketchy information given. Sometimes the selection is easy, when you see a major company wants to talk to you.Among several offers, there was one from Canonical, which is the firm supporting the Ubuntu operating system for desktop, server, and commercial uses. I took the opportunity without hesitation.

At ARM TechCon, Canonical issued a press release, Canonical joins Linaro Enterprise Group (LEG) and commits Ubuntu Hyperscale Availability for ARM V8 in 2013. I sat down with Michael Kress, VP of Sales at Canonical, and talked with him about it.

Michael Kress

Michael flew in the night before, after waiting at Logan airport for four hours. He was flying from Boston, which was impacted by our friend Sandy the superstorm. He has been with Canonical for close to five years and has worked in several areas, including the ecosystem of ARM for Canonical. I did not know much about Linaro. According to Michael, Linaro was started about two years ago with four to five companies. I was led to believe that Linaro is for embedded Linux, but Michael corrected me. The embedded Linux world is quite different from that of non-embedded one, and it is not easy to enter it without the right background.

Linaro's activity is summarized from their website:

Linaro is the place where engineers from the world's leading technology companies define the future of Linux on ARM. The companyis a not-for-profit engineering organization with over 120 engineers working on consolidating and optimizing open source software for the ARM architecture, including the GCC toolchain, the Linux kernel, ARM power management, graphics and multimedia interfaces.

To ensure commercial quality software, Linaro's work includes comprehensive test and validation on member hardware platforms. The full scope of Linaro's engineering work is open to all online. Open engineering has been practiced from the start at Linaro with plans, specifications and progress available for inspection on the developer Wiki. Linaro is distribution neutral: it wants to provide the best software foundations to everyone, and to reduce non-differentiating and costly low level fragmentation.

Linaro Enterprise Group (LEG) is a new working group at Linaro that includes companies like AMD, AppliedMicro, Calxeda, Canonical, Cavium, Facebook, HP, Marvell, and Red Hat, in additon to the original Linaro members, like ARM, HiSilicon, Samsung, and ST-Ericsson. Linaro released information here about it.

Canonical/Ubuntu is embracing ARM. Because my areas of coverage have grown so wide, I was blind-sided about Ubuntu's new territory. But when you see the growing market into SNS, it is no wonder that Ubuntu is entering the ARM world. In the SNS world, what is required is scale-out but not scale-up. Computing tends to be distributed, and not every task requires heavy computing. For that, a server based on ARM is a good fit. Canonical/Ubuntu was hinting at this in one of the blogs by Victor Palau back in August. In the blog, there was a link to his presentation to summarize why. I thought it very useful. In the presentation, Victor showed some interesting numbers, such as:

• 98% of 1 billion mobile phones sold each year use at least one ARM processor.

• A rack designed for 40 traditional servers could house 3,000 ARM servers.

• HP estimates ARM servers will require 94% less space.

There was more stuff in the presentation, but let me defer it to a future blog.

Canonical started with ARM in 2008. In the actual working with Linaro, Canonical assigned a few engineers dedicated to Linaro to work with other engineers from other companies to solve tough engineering problems, such as the interoperability of various technologies and boot-process differences among system-on-a-chip (SOC) vendors. There are several packages of Ubuntu tailored to each platform, and each version is natively compiled. For an outsider like me, it would be very interesting to compare benchmarks between their X86 version and the ARM version, but for an obvious reason they keep such benchmarks internal. The next question is whether an ARM-based server is ready for prime time. Michael thinks the 32-bit ARM solution is ready for specific application areas like Web and Big Data, which require scale-out features. Canonical can expand into many areas, such as cloud computing, but it does not work in the embedded market. In the embedded market, an OS is sometimes not a visible component.

From the green IT perspective, this is interesting. It is welcome news to know that both Ubuntu and Red Hat are embracing low-power servers based on ARM chips. If a low-power server running ARM replaces one with X86, power consumption goes down tremendously. But if you pack more servers into the same or less space, what happens to power consumption and heat emission? Take a rack of servers. With ARM servers, the density would be 75 times (according to Victor above) greater than the traditional way. Compacting the servers would minimize space, and that would reduce energy consumption for sure. That is welcome news. On the other hand, if you increase density and pack more into a given space, power consumption and heat emission may increase. For that, alternative cooling technologies and operational methods may be required. I have not seen any research on the tradeoffs anywhere yet. At this time, examples of ARM servers in many data centers are hard to find, and I defer my comment on this to a later time.

At the ARM TechCon, I interviewed four companies that contribute to low-power computing from the perspective of various kinds of software. IS2T is based in France and provides very small footprint Java development and execution platforms. They announced two new products at the show. One is called MicroEJ and the other is MicroJvm Virtual machine . Java environments exist for enterprise (standard edition, SE) and embedded systems (micro-edition, ME). Java ME is further divided into Connected Device Configuration (CDC) and Connected Limited Device Configuration (CLDC) flavors. IS2T supports Java ME with the CLDC flavor that is suitable for very low power computing.

The spokesperson I interviewed was Regis Latawiec. Although his title is executive VP of business development, he was very technical and I enjoyed a conversation with him. Technical people tend to get enthusiastic when they meet with someone who can appreciate and understand their technologies. Well, I am not an expert but someone who knows just enough to be dangerous. In the past, I dealt with Java ME at a startup that developed and marketed security solutions for a card reader. Increasingly, cards like credit cards incorporate a smart chip that cannot be tampered with, but a card reader may not be very secure. The startup’s product was to provide a security layer, similar to the mobile information device profile (MIDP), for card readers. I am by no means a Java developer or an expert in the subject. Back to IS2T. While I was reading their press release, I had several questions and this was a good venue in which to clarify my questions. See, I am not a reporter who simply takes what was said and writes an article about it. Instead, I analyze it and express my opinion.

Regis Latawiec

MicroEJ: MicroEJ, Java development platform, is a product that consists of many elements, such as SDK, JVM, and a simulator. It uses open source Eclipse as an SDK base and expands its functionality. Because Eclipse supports multiple languages, with plug-ins, the SDK can support C and C++, which are still the dominant development languages; there are many legacy C and C++ applications still in the embedded segment.

The following is a screenshot of the SDK function of MicroEJ.

MicroJvm Virtual machine : Their JVM requires only 28 KB of flash memory and less than 1.5 KB of RAM. It also boots quickly, in 2 ms at 120 MHz. They estimate that a fully functional advanced human machine interface (HMI) requires 90 to 140 KB. By comparison, a solution from an incumbent requires 350 KB of ROM and 130 KB of RAM for CLDC.

Simulator: We can take two different approaches in development. One is to settle on a hardware platform first and then develop software for it. The other is to develop the software first and then find a suitable hardware platform. In many cases, the hardware platform is a given. Because a typical embedded system is resource constrained, we do not develop software on the target platform. Instead, we develop it on a much more powerful platform, like a PC with a faster and more powerful processor and abundant memory. But the downside of that is the need to cross-compile and download the binary to the target machine for testing. If the tests fail, we need to go back to the development environment and fix the code, then repeat the process until the code is perceived as acceptable.

With a simulator, we can develop and test software on the same platform, saving effort and the time needed to go back and forth between the target and development platforms. But it is more than effort and time; it requires extra power to do so. From the green IT perspective, this saves extra energy. Better yet, this is Java and "develop it once and run it everywhere.” So we can develop and debug software on the development platform. The very code we developed on the development platform should run on the target platform without any change. Also, no cross-compiling is necessary.

What about performance? In general, a simulator does not simulate performance, for the obvious reason. However, IS2T's simulator provides information, such as task scheduling and memory requirements. If we have the latitude to select or adjust the target platform, we could choose the best one on the basis of the information. If we do not have that freedom, we could use the information to improve the software. Experienced programmers have a sense of how much memory is required for some particular function. If they have the memory requirement information for the current version of the software, they can use that information to fine-tune the code to reduce the memory requirement.

Legacy applications written in C and C++: C and C++ applications are dominant in the embedded world. If you provide only Java solutions and ask users to switch their programming languages to Java, you would not win them. So what IS2T did was to develop a Java platform that interfaces with those legacy applications. Java Native Interface (JNI) is a standard that allows a Java program to call programs in both C and C++ and vice versa. Their MicroJvm Virtual machine can run on either RTOS or bare metal, maximizing flexibility. Regis showed me a sophisticated diagram that was too detailed for me, so I drew the following, simpler diagram. MicroJvm Virtual machine allows the reuse of legacy programs and shortens deployment.

Support for legacy applications in C and C++.

But that is not all. From the green IT perspective, this is very energy efficient. I mean each application, regardless of which programming language it is written in, consumes a lot of energy when it is developed. Software development means human activity. Programmers need to be housed somewhere, and that alone consumes energy with lighting and temperature control. Software needs to be developed on some kind of computers, and they consume energy. Source code may have to be printed out for detailed discussions, which consumes energy. Programmers may have a coffee pot for relaxation, which costs energy. If we throw them away and start from scratch in Java, the resultant system may be cleaner (in terms of logic but not of greenness) and more consistent. But additional energy must be consumed for that.

Future: The Internet of things includes many objects with networking capability. Some do not interact with humans but others do. If such an object is resource constrained but needs a human machine interface, that HMI will be very primitive because a sophisticated interface requires a certain amount of resources. With something like MicroJvm Virtual machine, many such objects could be installed with more sophisticated HMI.

Competitive edge: Finally, I asked if they had some patents to show their technical superiority. Regis’s answer was very interesting. If they apply for a patent and get it, some of the details of their technologies may be disclosed. It is their decision to keep their technologies secret. Then how do you know if they are really that good? He said he would show it by real examples. Give them a week for a fairly sophisticated specification, and they will deliver it in a week. That is proof! Also, their secrets are declared (in secret) to a lawyer organization so they can prove their invention. It is very different from what I know, but it is interesting.

ARM spec chips are known to consume less power and are suitable for resource-constrained applications. At the recent ARM TechCon, the main theme was low-power computing. But a chip alone cannot accomplish that. Low power means less computing power and less memory and storage. Utility, middleware, and application software must be designed for and tailored to that kind of environment. I interviewed four software companies at the conference and one of them was IAR Systems. After all, I am a software guy. And one thing I like about embedded systems is that they use C and C++ predominantly, with some Java, which I can still read (but not write anymore).

I sat down with some IAR folks to find out more about their power debugging. When I interview a company, it is usually one or two people who come to see me. This time it was a bargain. I got four of them who cover CEO, marketing, technology to international market functions as a group. Because of this, I got a pretty good insight into the company and its products.

From left: Kiyofumi Kamimura, Thomas Sporrong (standing), Fredrik Medin, and Stefan Skarin.

Company

IAR is headquartered in Uppsala, Sweden. This city sounds familiar because MySQL was headquartered there before its move to the US. Kiyofumi is a country manager for Japan. We could have used Swedish, English, or Japanese, but because my Swedish is rusty, they were nice enough to agree to conduct an interview in English.

IAR has been around for 30 years, so we cannot call it a startup anymore, for sure. The total number of employees is around 160 and 70 of them are developers. Turnover is extremely low. Over the years, only three people have left the engineering development team. Most of the developers are gray haired after many years of experience with the company. Some have been with the company for thirty, twenty, or fifteen years. They are still doing basically the same job. It is quite a contrast to Silicon Valley. Large Japanese companies are known to have very low turnover, although that is rapidly changing. Even though people may not leave their company, they move around within the company in Japan. If you are a programmer and with your company for decades, you tend to be promoted to manager and no longer touch programming.

This is one of the strong points of IAR. Their clients are happy because the same guy works on their project for many years and it is easy to communicate with the programmers. Also, internally useful know-how and knowledge of how to produce efficient and high-quality code accumulate over the years. That is a great asset for them. They focus on delivering products but not services. Thus, this vast amount of software development knowledge is kept inside the company.

Power Debugging

From the green IT point of view, measuring hardware energy consumption and comparing its energy efficiency with that of other hardware are fairly straightforward. We could measure power consumption and compare multiple hardware platforms based on the measurement. What we do is to put a probe into a circuit board and measure signals and power. But when it comes to software, it gets more complex and difficult. Given a specification, there are many more ways to implement a solution with software than with hardware platforms. Even using the same algorithm, we could still fine-tune it and adjust data structures. For software performance, we have used something like profiling. I was a software programmer once and used it myself. By now my programming skill is so rusty that I do not want to mention that I earned my living by writing programs in the past.

The difficulty of software is code itself. Code is a static entity, and it is very hard to predict software behavior in action. That is why profiling is very effective. We can gain dynamic behavioral information, such as how many times a particular section is exercised in a run. With that information, we could modify and improve software to run more efficiently. Although that is very useful information, it is still logical information. Profiling is mainly used to speed up software but not necessarily to reduce power consumption. If we are on a platform that has plenty of computing power, memory, and storage, then speed is very important. But in a resource-constrained, embedded world, sometimes power consumption reduction may be more important than speed.

What if we can get more precise information about what each section or statement of software consumes in terms of electrical power? That would be very helpful for programmers in fine-tuning their software, and it is what power debugging is all about. As shown in the picture below, as in a typical debugger you can see the program execution at the source code level along with the assembly code equivalent and power consumption information at the statement level. By observing how power consumption changes as the execution proceeds and how much power each statement consumes, the programmer can modify the code to reduce power consumption. Of course, that information alone would not show how to change the code. As you know, a software modification tends not to be localized except in very simple cases. One change may have multiple impacts within the same function but sometimes also in other functions. That is up to the programmer.

The overall system power consumption is shown as a graph in the Timeline Window. Double-clicking on any part of that graph will take you to the corresponding location in C source and disassembly views above.

The setup seems to be fairly straightforward and easy. What you need is a circuit board, power debugger probe (I-jet), and GUI software on a PC, as in the picture below. I-jet has an interface to a circuit board and provides power to the board via a host computer, the PC. I-jet is connected to the PC via USB. By the way, IAR's product works with C and C++ code, which is dominant in the embedded world.

GUI on a PC connected to a circuit board via I-jet (yellow box).

In addition to tuning software performance, a power debugger like this can be used to fine-tune a circuit board. If we find higher power consumption than expected, we may want to lower clock speed or reduce the RAM size, sacrificing speed. On the other hand, if we find that the program does not require as much memory as installed, we could replace it with less memory.

Future

This is very interesting. What will they do next? They told me that the current version allows putting a probe into the entire circuit board. The future version should allow measuring only a particular micro-controller rather than the whole board.

Speaking of the future, I asked them about the applicability of their solution to the server environment in a data center. Most attention has been given to the facilities and hardware side of IT for energy efficiency. Even power usage effectiveness (PUE), the most used data center metric for energy efficiency, only considers the total power consumption of the data center as a whole and IT as a whole. People are starting to shift their focus to software for energy efficiency. It is far more difficult to deal with software because it is more complex than hardware.

They are currently focusing on embedded systems and pointed out a couple of things to me. One is that in the embedded world, each software is pretty unique and developers probably need to write 90% of the code, with the rest coming from libraries and such. So power debugging makes sense. In contrast, in the enterprise environment, programming is minimal and the rest is done by integrating existing code. Because of that, its relevance in the enterprise environment may be minimal. That may be true, but we can use power debugging to pick the most energy efficient libraries and utilities/middleware when we have the freedom to choose. The other point is that ARM is entering the server world, in addition to the embedded world. For example, AMD, HP, and others are planning ARM-based data centers. AMD acquired SeaMicro, an ARM-based server vendor, earlier this year. HP and Calxeda are working on an ARM-based server named MoonShot.

If ARM-based chips become prevalent in the data center environment, many software packages will have to be rewritten to run on the new ARM platform. If that happens, IAR's power debugging idea can be used.

James Urquhart

There have been a lot of talks on cloud computing, about what it is, what kinds exist, and how it changes the way IT is delivered. Virtualization, security, automation, provisioning, tools, and other details have been discussed many times by many people. But his talk was the first I’ve heard that specifically covered how to design the entire system of cloud as a whole.

If I may oversimplify his point:

• Cloud is an interconnected complex adaptive system.

• The interconnected system as a whole may collapse, even if we carefully design and operate each component of the system.

• It is vital to build resilience into the design and operation of such a system.

He gave an example of a market crash, known as Flash Crash, which occurred on March 6, 2010. I myself did not remember it, and neither did most of the audience. An automated system for stock exchange executed the sell program extremely rapidly in just 20 minutes. With this, the market got more active and more orders poured in. The system spun more threads to process these orders, and finally it got out of hand and crashed. With this story, James introduced the term complex adaptive system.

It is defined as:

Further,

where emergent behavior or emergence is defined as:

the way complex systems and patterns arise out of a multiplicity of relatively simple interactions.

Because each component of the complex system is relatively simple and easy to control, we tend to think we can control the system as a whole. But as the discussion above shows, that is not the case. If there is no way to predict future behavior, the only thing to do is to accept potential failure and get ready for it. Because there is no way of knowing how the system may fail, we may randomly produce a large number of problems to test our systems. James showed one such system that causes random chaos, Netflix's chaos monkey.

I usually relate the subject of this blog to green IT or energy. That’s hard to do in this case. One thing I can say is this: if the complex adaptive system fails and starts to behave erratically, that surely wastes energy, because the behavior does not produce any meaningful outcome. But I need to think further about what this means to green IT.

If you want to know more about it, you can read some of his blogs.

• Exploring cloud and complexsystems

• Five resources for learning complex adaptivesystems

• The art of failure in cloud’s complexsystem

James recommended reading Drift into Failure by Sidney Dekker of Griffith University, Australia, and other books, shown below, to further understand this.

The first company I talked with was yaSSL ("yet another SSL”). I wrote about them some time ago. The last time I talked to Larry Stefonic, their cofounder, I wrote about their products and features. This time I want to focus on their origin and the market segment they serve, low-power computing.

As I wrote before, Larry was Senior VP of sales at MySQL, which was successfully acquired by Sun (and then became part of Oracle). The following is a summary of my conversation with Larry.

Larry Stefonic

Origin

Back in 2005, MySQL did a code audit to make sure that all the code in MySQL could be verified as legitimate. They wanted to make sure that nothing in the MySQL code violated any ownership claims from others. It was important for MySQL to make sure that MySQL products, regardless of whether these were open-source or licensed versions, did not violate any license requirements from others, so that their users could use their products without worry of any claims.

The audit revealed that the OpenSSL package they were using had two problems. One was that it had grown too big and become unmaintainable. OpenSSL was also unworkable because it has a complex license and unclear copyright ownership. For the latter, it was hard to continue with it when MySQL was delivering commercial licenses to its code base. Hence, a new SSL package was necessary, which is open source and works for the dual-license environment. That was the starting point of yaSSL's birth.

Merits

Well, SSL is standardized. Then how do you differentiate yourself from the rest of the SSL competition? After all, anyone with a reasonable background could develop another SSL package. Larry listed the following points of differentiation:

1. The Open-source business model keeps the costs low; and that benefit is passed along to commercial users for as little as $5,000 for a single product license.

2. Small footprints with few resource requirements that can run on embedded systems, even on bare metal. The code size can be 20–100 kB and the run-time memory requirement is 5–50 kB.

3. Super portable.

4. Solid commercial license option.

5. Hardware-based crypto, including Intel's AES-NI and Cavium crypto processor.

Many of the above came from the belief that the embedded-system market would grow and many embedded systems in the future would be networked. If those resource-hungry embedded systems are on the network, they would certainly need encryption. If that is the case, no need to explain the aforementioned merits.

Markets

Now, where are yaSSL packages suitable? Larry listed the following markets as of interest:

1. Streaming media (video and voice) over the Internet, leading to support for DTLS, streaming ciphers, and TLS 1.2. For example, Skype uses the CyaSSL package.

2. Sensor applications in power grid, such as meters, and connected home appliances, such as those of Whirlpool and Bosch.

3. Gaming applications. CyaSSL was ported to all the major gaming consoles, such as the PS Vita console on SONY's Wipeout 1214.

This is very interesting. It is necessary to downsize everything to make embedded systems ubiquitously connected via the Internet.

While they were in town, I arranged for them to talk to data center experts, including those at Vantage Data Center. The best way to do that is to tour their facility and talk with the knowledgeable folks there. The JDCC folks had a good time in both the tour and the discussion. I visited Vantage Data Centers before and was impressed with their adoption of new technologies and operational methods. The Vantage campus is in Santa Clara, CA, and consists of three buildings, V1, V2, and V3. Intel had a manufacturing operation there when it owned the site, so some necessary infrastructure was already available. In November 2010, I attended a data center event in the building that is now named V1. On my last visit, I toured V3. At that time, V2 was under construction, V1 was being renovated, and V3 was partially occupied. This time V2 was completed and occupied. When I last visited, V2 was submitted to the USGBC for LEED Platinum certification, and V3 was certified for LEED Platinum. Now V2 is also certified as LEED Platinum, as the sign on the wall indicates in the following picture. It is very interesting to see how a data center goes through its development phases.

Front row from left: Hideki Okita (Hitachi), Leah Lovelace, Atsushi Yamanaka (IDC Frontier), and Rezin Brannon

Back row from left: Jun Sato (Mitsubishi), Minesh Patel, Christine Zinkgraf, and Justin Thomas.

In addition to touring V2, we exchanged information in a lively discussion. On a typical tour you tend to receive information but are not able to provide any meaningful information back. But in this meeting, some concrete information about data center operational methods and technologies was exchanged. I hope the Vantage people gained some useful information, almost nonexistent in English, about Japanese data centers. By the way, the following figure summarizes the highlights of V2. Other details are given here. The general information on their data centers, along with V1 and V3, is available here.

Highlights of V2 data center

The remainder of this blog summarizes the discussion between Vantage and the JDCC.

Locations/siting: In the US, data centers are everywhere, in urban, suburban, and rural parts of the country. In addition to its site in Santa Clara, Vantage is constructing a new data center in Quincy, in central Washington State along the Columbia River, where hydropower is abundant and inexpensive. Dell, Microsoft, Google, and Yahoo have data centers in the same vicinity. As is often discussed, some of the major factors in siting a data center are (1) inexpensive and ample power, (2) fiber availability, and (3) market. Silicon Valley satisfies (2) and (3) for sure, and (1) partially, with Silicon Valley Power. On the other hand, central Washington satisfies (1) and (2).

Vantage has two dedicated substations (50 MW each) and is a carrier neutral for providers with over 10 coming into their data center in Santa Clara, so they easily satisfy power and fiber requirements. On top of that, it is in Santa Clara, the heart of Silicon Valley, and has access to the Valley market.

In Japan, most (70%) data centers are in Tokyo and the surrounding area. Hitachi’s data centers are mostly in urban areas, and Mitsubishi has only urban data centers. Softbank IDC Frontier had only urban data centers before but recently constructed new ones south and north of Tokyo in nonurban settings. Their list can be found here. People in the US tend to think Japanese operators look for a place where power cost is the lowest. However, there is very little difference in the price of power throughout the country. Before the disaster, the only siting criterion was to be situated in the urban Tokyo area. Most major companies' headquarters are concentrated in Tokyo, and server hugging is given as another reason for that.

Power consumption: We can talk about power consumption in terms of kW per square foot or kW per rack. Because of the metric system used in Japan, kW per square foot was not a good unit to use in our discussion. IDC Frontier provides a colo service and its rack density averages around 8 kW/rack, but it does not have anything like 20 kW/rack. Vantage's power density varies from 3–5 kW/rack to 12–15 kW/rack.

Power supply/delivery: Vantage's power system is 2N and the data centers have two distinct power supply paths for redundancy. This satisfies Uptime's Tier 4 definition from that perspective, but they do not meet the other requirements for Tier 4. So this may be classified as Tier 4 minus. The data center under construction in Quincy qualifies as Uptime Tier 3. By the way, the JDCC felt that they needed a guideline that is similar to Uptime's but more reflective of the Japanese environment, which includes fewer power outages and more active seismic conditions. So they have created their own.

UPS equipment: Vantage listed MGE/Schneider Toshiba as UPS vendors. The first is utilized in their legacy spaces and the latter is used in their new energy efficient designs. The JDCC said that Japanese operators prefer Japanese vendors like Mitsubishi and Hitachi, although APC/MGE/Schneider has a presence in Japan. The reason for the preference is the lack of solid technical support. The offices of foreign vendors in Japan are more like sales offices without competent technical staff who can service their products in Japanese.

Cooling: Vantage uses air economizer extensively. When the temperature is below 95°F, ambient air is used for free cooling, which is more than 80% of the time in Santa Clara. Air economizer was not used in Japan before, but some data centers built in the past few years, like the ones by IDC Frontier, are now equipped with this feature. The conventional raised floor becomes ineffective when power density exceeds more than several kW/rack. In Japan, most data centers still use raised floors for cooling, but some operators, like IDC Frontier, provide both the raised and the slab floor with hot-air containment.

LEED and LEED equivalents: While EPA's data center EnergyStar rating is taking time to sink in, LEED has become the de facto standard for data center energy efficiency. Vantage makes an extra effort to get certified with LEED. Each country has its own version of such certification. In Japan, Comprehensive Assessment System for Building Environmental Efficiency (CASBEE) is used for certification.

I am sure I missed some other interesting discussions, but overall it was a good opportunity for the JDCC to see the data center that reflects the current trends of technologies and operations. Thanks to Vantage for the opportunity!

 Although I was involved in software development before and touched the embedded world a little bit, I am not an expert in the field. However, when I found that Prof. Jonathan Koomey was one of the keynote speakers in the conference, I decided to attend. He is a well-known researcher into the energy used by computing. One of his works was mentioned and used in the 2007 EPA report on power usage by US data centers.

 Prof. Jonathan Koomey

His point was to show how computing has improved in terms of the number of computations per kWh. This is defined as:

       [number of computations per hour at full load] /

                                        [measured electricity consumption per hour  at full load (kWh)]

This is only for the full load. As there were a few questions from the audience, he did not measure anything other than the full load.

In contrast to this measure, Moore's law states:

       the number of transistors on integrated circuits doubles approximately every two years

Koomey's metric may not cover all cases, but it is a little bit more scientific than Moore's. With this metric, he analyzed the historical data and concluded that the doubling time for performance per computer was about 1.5 years in the PC era. This means that every 1.5 years performance got doubled. This translates to 100 times better performance per decade, from the 1940s to the present. This ushered us into the laptop/tablet/mobile era, then to the age of the likes of low-power computing and sensors. In the latter, low power is more important than high efficiency. He showed a few interesting examples of devices that harvest necessary power from digestive fluid (in the case of a tiny device embedded into a medicine capsule) and stranded electric signals, such as TV/radio broadcast, wireless waves, cellular waves, and other electric magnetic waves (this is in urban areas only). Other sources of power for these new tiny devices include heat, light, motion, and blood sugar. Of course, personal computing will not go away, but this is exciting.

Now the next question is how far can we push this trend? What is the limitation? Is sky the limit? Koomey referred to Prof. Richard Feynman's work and said that with this trend, the theoretical limit would not be reached until the year 2041. That is another 30 years! In the development of processors, we seemed to have hit the limit of the number of transistors we can put on them, because of the heat-emission problem. But that is not the theoretical limit discussed here. There will be a different way to diffuse it, and this theoretical limit still holds. How much more progress will we make until then? I do not think I will be around to see that, but it is really intriguing.

He summarized his talk in the following slide.

In the opening speech of the 5th SVLG Data Center Energy Summit, Anne Smart, Director of Energy, Silicon Valley Leadership Group (SVLG), gave a good introduction to this topic as well as to Andrew Feldman, Corporate VP and General Manager, AMD, who delivered the keynote speech. You can watch their speeches here (a little over 16 minutes).

Andrew was CEO of SeaMicro, which developed and marketed low-power servers and was acquired by AMD earlier this year. Because the video is less than 20 minutes, you may want to invest your time in watching it yourself. But his point was simple. There is an area where we made significant progress in energy efficiency for data centers, but there is another area where we did not make much progress. The former is the facility side and the latter is the IT side. A lot has been done about energy efficiency in the facility area, such as finding better ways of cooling (hot/cold aisle and air economizer), which Andrew called low-hanging fruit. Now cooling efficiency has improved to three one-hundredths of the power required to run a server.

Now about IT. Let me inject my thoughts here before going on with Andrew’s speech. Before he started discussing the energy efficiency of IT in the data center, I knew I was going to agree with him. We have been trying to alleviate the symptom of data center energy problems by attempting to control cooling and power rather than curing the root cause of the data center energy crisis, that is, efficient ICT equipment and making better use of it, that is, running it without utilizing its full capacity. Emerson in its energy logic claims that IT is the root cause of the large consumption of power in data centers and we had better control IT. We are beginning to pay more attention to IT as we try to control data center power usage. We need to pay more attention to using more energy efficient equipment, such as SeaMicro's, and using virtualization to increase the utilization of each server. We need more than PUE to gauge and measure IT efficiency. There have been a few data center metrics proposed, such as McKinsey's CADE. The Japanese have proposed DPPE. Both metrics take IT’s behaviors into consideration. I recently found an interesting metric for IT energy efficiency called Par 4. I have not reviewed it yet, but the short description I read sounds interesting. I plan to write a separate blog on it.

Andrew started to talk about IT (servers) in terms of data center energy efficiency. He dealt with server energy efficiency at SeaMicro with low-power chips before he came to AMD. And he is one of the most qualified persons to discuss the subject. He bluntly stated that he was disappointed at the progress we have made in that department, because a server is the most powerful consumer of power in a data center and needs more attention.

He mentioned that like anything else energy efficiency improvement for servers began with the low-hanging fruit. In this case, it was the power supply. The power supply used to lose 18% in transit from PDU to server. Now with advances in power supply technology, it is down to 7%. But the real meat in energy efficiency is the server itself. At SeaMicro, Andrew and his team developed a server that requires only a quarter of the power necessary for other servers. But it took five years and $50M, which was not readily available from the VC community in the funding climate then. He is thankful to the Department of Energy and the State of California for their grants that made it possible for SeaMicro to complete their servers.

He wrapped up his talk by saying that we should not stop innovating in energy efficiency in both facility and server technologies. Power is such a big problem in the data center that we need to keep working on it.

Atsushi Yamanaka of Japan Data Center Council (JDCC) talked about:

• What happened to data centers at the time of the major earthquake in March 2011

• What data centers are like a year and a half after the disaster

Atsushi Yamanaka

Atsushi and I teamed up in a session last June at a different data center conference, and I think he delivered his talk even better this time. London-educated, he is an excellent speaker. His profound data center knowledge, in both technologies and business, sounded very convincing in his British accent. And no wonder, because his day job is general manager at the president’s office at IDC Frontier.

 I will not recap all of his talk here but some of the information I found most interesting:

• Some rumors and hearsay about damages to data centers in Japan

• Current data center trends in Japan

After the earthquake and tsunami, rumor had it that all the major data centers had been damaged or washed away by the tsunami. Also, radiation was believed to be so bad that people were running away from Tokyo. I have been to Japan probably more than ten times since the disaster, but in Tokyo it has been business as usual. There were a few signs of the power shortage in 2011, but in 2012 I no longer see them. But don’t take my word for it; listen to what Atsushi said. He told us that data centers are concentrated in Tokyo (70%), with 20% in Osaka and the remaining 10% scattered around. Tokyo is 140 miles from the damaged nuclear reactors in the Fukushima-Daiichi power plant, far enough not to have felt any impact. Yes, the shaking was tremendous and lasted longer (especially in high-rise buildings) than most people had experienced before. But all of the data centers in Tokyo were unscratched or had minor damage, like cracked floors.

Atsushi’s presentation included a video shot during the quake that showed a guy being shaken badly as he serviced one of the racks. But all the racks had been bolted to the floor, and none of them fell down.

Atsushi also said that energy efficiency was in the blood of the Japanese people before the quake, and now they are even more concerned with saving energy, including power. Many new data centers were developed with that in mind, including those at IDC’s Kitakyusyu site, Sakura net, and Nomura Research Institute. Modularity and air economizer are becoming norms in the construction of new data centers. Those who want to find out more can ask for more information at info at jdcc dot or dot jp (change "at” and "dot” with appropirate Internet formats).

The cost of power varies drastically by location in the US, so one of the important elements in siting a new data center is power cost. The price variance is very small throughout Japan, and power quality is probably pretty uniform. The reason for the 70% concentration of data centers in Tokyo is because the city is the nation’s center of business, politics, and academia. That is changing a little, but it can accelerate still more.

We were fortunate to have with us two additional members of the JDCC:

• Jun Sato (Mitsubishi)

• Hideki Okita (Hitachi)

 Atsushi and those guys compete daily but unite for the sake of the data center industry in Japan. They added some perspectives from traditional and large enterprises in Japan.

From left: Atsushi Yamanaka, Jun Sato, and Hideki Okita

The crowd at the session. After this photo was taken, more people came in, and the room was packed, with lots of people standing.

I do not remember all the questions the attendees asked, but a few stuck in my mind. Some wanted to know if the US (or Silicon Valley) is prepared for earthquakes. Hideki was polite but I am not. I do not think the Bay Area is ready. I think data centers here need more preparation.

The Silicon Valley Leadership Group (SVLG) indicated that they would like to invite the JDCC next year as well. OK, JDCC guys, next year you need to top what you just did!

The main speaker will be Atsushi Yamanaka, who heads the International working group at JDCC. His day job is at IDC Frontier. He informed me of the three main topics of his talk, and I would like to share them with you briefly here:

1. Current data center distribution in Japan and how each region coped with rolling blackouts immediately following the March 2011 disaster.

2. Various methods for efficient power consumption actually practiced by data centers in Japan that are coping with power shortages, together with a new efficiency measure, DPPE, begun to be used by data centers.

3. How the Japanese data center industry united under the JDCC to solve industry issues such as power reduction mandates and greenhouse gas reduction mandates without sacrificing business growth potential, including new industry guidelines for power efficiency practices like shutting down servers, raising data center temperature, handling standby servers, implementing cold or hot aisle containment, and enhancing heat exchange in outside units of cooling systems.

I also asked Atsushi to give us a brief introduction to the JDCC and what they do. These are pretty interesting points and you may want to come and join me to find out more details. After his speech, I will add panelists from NEC, Hitachi, and Mitsubishi who are data center operators and vendors themselves and get their input.

This is a good time to find out what's really happening in Japan! Don't miss it.

Yes, in the US we experience a power availability and supply problem when we try to build a data center or expand capacity. Some operators try their best to maximize available power for expanding their existing data centers, while others look for a building site with an ample power supply. Recently, the New York Times ran an article critical of data centers’ heavy power consumption. Does the data center industry do nothing but waste precious power? The answer is NO! Silicon Valley Leadership Group's (SVLG) Data Center Energy Efficiency (DCEE) Summit is the venue where energy efficiency is extensively covered and discussed.

Anne Smart, director of energy at SVLG, is quoted in the following excerpt from a Politico article:

"There is no denying that we're increasing our need for data centers.” What the Times story got right is the rapid expansion of the data center industry, Smart said. But the article shied away from technologies people are working on to help with data center efficiencies. "As the community looks at new technology, data center operators are incredibly open to figuring out best practices and working collaboratively to address the energy demand of data centers,” she said.

On October 24, the Silicon Valley Leadership Group, in partnership with the California Energy Commission and Lawrence Berkeley National Lab, will host its 5th Annual Data Center Efficiency Summit. PG&E is a key supporter of the event, which will be held at Advanced Micro Devices' California headquarters. One highlight—The Japan Data Center Council will discuss what they learned about operating within energy limits after last year's tsunami and earthquake.

After the disastrous earthquake in Japan and the tsunami that followed it destroyed four nuclear reactors, the rest of the reactors were stopped for fear of similar accidents. This caused a power shortage on a national scale. Before that, Japan had been proud of its solid power grid infrastructure and ample power supply. Even in such an environment, data center operators there tried their best to maximize power use. But now, in the face of a national power shortage, Japanese data centers sometimes face criticism similar to that in the Times article.

Some ironic trend is causing Japanese data centers’ power demand to increase. There were pockets of server rooms and closets scattered around the enterprise before the disaster. But because of the unstable power supply, those servers hidden in small pockets of buildings are being moved to data centers for a stable operating environment. Although no immediate rolling blackouts—exercised right after the disaster—are eminent, many data center operators are worried about the possibility, as the power supply to some data centers was shut down for two hours at a time before. Many data centers are equipped with a backup power supply, from diesel generators, that lasts only 48 hours at best. Technically, this allows them to survive 24 two-hour blackouts. But diesel generators must eventually be refueled. Further disasters might interrupt the smooth flow of refueling—delivery by truck might be interrupted by traffic congestion and damage to the traffic infrastructure.

The Japan Data Center Council (JDCC), a consortium of data center operators that boasts more than 100 members, is sending a delegate to present at the SVLG DCEE conference and to talk to data center operators and experts in Silicon Valley to find out what new technologies might help them cope with power problems. I will moderate a special session given by Atsushi Yamanaka and other JDCC members. Come and meet us, and find out how they are operating their data centers in such a constrained power environment.

Justin Rattner started his speech wearing the latest communications gear with moving ears.

The theme of his speech was very clear, as the following slide shows.

It has been said that connected devices will be everywhere, comprising the Internet of things. But when the Intel CTO spoke, that statement carried weight. I attend many conferences and take pictures for my blog, but usually only a few people take pictures of speakers and their slides. In this case, a bunch of people were taking pictures every time he advanced his slides.

Intel's main business is in semiconductor chips, but it has a wide variety of technologies to push its vision for connected devices. I am sure there are many more technologies in development at Intel, but Rattner demonstrated six. He reminded us of what was talked of as a dream at IDF 2002, 10 years ago. At that conference, someone from Intel said that someday devices would be networked together wirelessly at a reasonable cost. With WiMax (Intel was a big proponent, but I do not hear much about it from Intel these days), Wi-Fi, LTE, and other wireless technologies, this is no longer a dream at all.

In any event, Rattner presented these six technologies:

1. Scaling down of communications chips. In the communications area, analog technologies have been in the mainstream with converters between them and digital technologies (computers are already digital). Because analog technologies are hard to scale down in size, a greater percentage of the communications parts are becoming digital. A chip once (circa 2002) manufactured with 90 nanometer (nm) technology is now (since 2010) done with 32 nm technology. With the scale down, size and power consumption are down from 1.2 mm² and 50 mW to 0.3 mm² and 21 mW. In two years or so (circa 2014), with 14 nm technology, its size is expected to go down to 0.03 mm². So scaling and power conservation are constantly being improved.
2. Wireless Gigabit (WiGig). WiGig Alliance's view is given as follows from their site.

Wireless Gigabit Alliance envisions a global wireless ecosystem of interoperable, high performance devices that work together seamlessly to connect people in the digital age. Our technology enables multi-gigabit-speed wireless communications among these devices and drives industry convergence to a single radio using the readily available, unlicensed 60 GHz spectrum.

WiGig Alliance’s president is also an employee of Intel. Multi-gigabit-speed wireless technology would enhance existing applications and make it possible to develop new applications.

3. Battery life is a concern for everyone. Batteries for mobile devices do not last very long, and until the technology advances to the point being able to store a lot more power, the best thing to do is to conserve power as much as possible. With its Smart Connect Technology, Intel has a NIC allow only absolutely necessary packets to reach the main computing engine, saving unnecessary processing and power.

4. Video Aware Wireless Networks is necessary, as video occupies a large portion of Internet traffic. From 2011 to 2016, its growth is expected to be 32% CAGR, and in 2016 video will be 55% of the total traffic, as shown in the following.

5. Security is necessary but usually diagonal to ease of use. They are advocating biometric technologies; Rattner’s demo used a palm for authentication. The technology, from Fujitsu, is used in several applications. Several years ago, Mitsubishi Tokyo UFJ bank adopted palm vein authentication for verifying each account holder. The Vantage data center uses it as well.

6. The wireless infrastructure needs to be improved. In the current implementation, each cell tower is self-serving and, because it is provisioned to accommodate the peak load, which may not occur often, it is also wasteful of ICT equipment/energy. By networking the cell towers together and controlling them from a central data center, load balancing can be applied and unnecessary equipment can be turned off without affecting the overall operation.

If you want to know more about Rattner’s presentation, I am sure you can find articles and blogs detailing each point he made. Instead I would like to ponder the role of ICT technologies in general, regarding their contributions to our society. Up until now, technical progress has been made to provide more convenience to us, as if we had an infinite amount of resources. It is only recently that we started to realize that we cannot continue to assume an infinite amount of resources.

In his presentation, Rattner did not state that his motivation was to conserve more power or energy. But in some cases, in order to keep up with demand, we must conserve power and energy and, on top of that, use what we have more effectively.

So looking at each technology area in his presentation from that perspective, area 1 concerns the need to make each component fit in a smaller area and consume less power. Areas 2 and 4 indicate that spectrum is a limited resource and should be used effectively to support loads. To do that, the technologies need to be improved. Area 3 indicates that, given limited battery capacity, the best thing to do is to conserve power. Area 5 may be the closest to the energy view. I think we can do a lot to make ICT technologies more effective in terms of energy consumption. And moreover, ICT can make other things more efficient and convenient. That is why I am interested in the field of green ICT.

Nati Shalom

Cloudify is a tool to smoothly launch applications in a cloud environment with a recipe that describes everything necessary, including resources and their configurations. GigaSpaces' new product (at least, I thought it was new) is eXtreme Application Platform (XAP), an accelerator for NoSQL databases. GigaSpaces' XAP is not a database, analytics tool, or visualization tool. In short, it is an in-memory utility to enable real-time data processing for other NoSQL databases, like Cassandra and MongoDB. SQL or no SQL, the rate and the speed of Big Data have become a problem for a database to process. A simple solution is to put some kind of front end in place to process such high-volume and high-speed data. In-memory data processing is usually much faster than any data storage dealing with disk I/O. Both VoltDB and Couchbase, which I interviewed at the same conference, use their implementation of an in-memory database for this. Other databases may partner with other companies to provide such a technology. Nati referred to Edd Dumbill's blog, which says that one of the trends in Big Data is streaming data processing. For that, in-memory technology is invaluable.

I thought XAP was a new product that came after Cloudify. However, XAP was developed about 10 years ago, when the Internet bubble was in full bloom. They thought high-speed data processing was necessary to accommodate business-to-business (B2B) interactions with scalability. As we all know, the dot-com era did not last very long, and their prediction did not materialize. Actually, the financial community was an early adopter because of real-time data processing in such things as credit card transaction processing and stock trading. So GigaSpaces decided to develop a product to serve those needs in 2004, and they have kept improving it over the years. The current version of XAP is the ninth edition. As I wrote in a previous blog, the NoSQL domain includes companies that develop databases, utilities, analytics engines, and visualization tools. This classification is shown in Matt Aslett's blog with leading NoSQL companies. Matt places GigaSpaces in the data/grid cache category.

The following is a summary of my chat with Nati about his solutions and his view of the NoSQL market.

GigaSpaces is headquartered in New York and also has offices in San Jose, CA, London, and Israel. Nati said that Big Data is fueled by different things, depending on the geography. In the US, SNS drives Big Data on the West Coast, while the financial requirements mentioned above drive it on the East Coast. SNS and financial applications are very different, but they both generate a high volume of data at high speed. SNS, especially, generates data in an unformatted way, such as tweets.

Regarding the relationship between XAP and Cloudify, they are currently tightly integrated. Data cluster management is necessary for the management of large data sets. Cloudify needs the same data cluster management for applications. Thus, the two share the same underlying data cluster management platform. After all, applications and data should go hand in hand for provisioning and management. Nati's blog describes this integration in more detail. In short, XAP accelerates data acquisition and Cloudify manages the cluster.

I was not sure about the relationship between the two products. Nati gave me a little more processing information, as follows:

When streaming processing is required, both XAP and Cloudify are deployed. If streaming is not required, Cloudify alone is appropriate. This is a logical diagram, but in reality those three boxes can run on a single physical server, or two on the same machine, because XAP should work closely with a database. Cloudify is written in Java, and XAP is written in both Java and C++. XAP not only accelerates data acquisition but also provides data processing and guarantees data consistency.

Next I asked him to draw a picture showing where something like GigaSpaces' XAP resides between NoSQL and NewSQL. Here's the picture, showing the two domains in an oversimplified manner for ease of understanding. Note that each database, whether NoSQL or NewSQL, is different in its offering and performance. For example, Couchbase claims high processing power, although it is classified as NoSQL database.

I asked him about the application of streaming data processing. Some utilities companies process Big Data with Hadoop to analyze meter-read data. Streaming processing may be a niche, but it is becoming necessary to process such things as meter-read data that may come in from millions of power meters in semi- or real time. It would be interesting to combine those data streams with weather data that might also change in real time. For a balancing authority like California ISO, which is tasked with balancing power demand and supply in real time, the real-time data sources vary and can be very large. It is necessary to source a large volume of data to process to get a good picture of the status of the power grid in real time to avoid blackouts. I have yet to see any examples of streaming data processing in the utilities business, but I think such an application area exists.

Nati mentioned that real-time requirements are growing and that Google, which invented the concept of Hadoop, is moving to Percolator, which supports real-time Big Data. Maybe this domain will not remain a niche for long.

The whole GigaSpaces system looks pretty complex, and integration seems to require a lot of hand-holding. Nati said that it normally does, but he makes extra efforts to make it very easy. He continued as follows:

"BigData systems are complex by definition - look at Hadoop, NoSQL, etc. What we do is integrate them in a consistent way and make reduce large part of the operational complexity and development complexity.

If you would compare the amount of effort that is required to build a twitter like real-time analytics with GigaSpaces you'll see that all you need to write few snippet of code to process your logic, scaling, fail-over, integration with BigData storage, management and monitoring is all curved out from the developers.”

They also provide training. One thing they thought of was an interface with popular NoSQL platforms like Cassandra and MongoDB. GigaSpaces has a semi-official partnership with those database companies. This is intended to exploit the fact that more people have worked with those databases; GigaSpaces can ride on their knowledge to lower the training curve.

Moving forward, I asked Nati to consult his crystal ball as to what will happen to the NoSQL/Big Data market. Will any standard emerge from a standards body or two? He told me that, as in many emerging markets, many of the companies will be consolidated and disappear, except for some like Hadoop, Cassandra, and MongoDB. As for the storage mechanism, one form is good for one thing but not other things. If there is a standard way of accessing data, key-value, tabular, or document-based data will be consolidated, but the forms themselves will survive because one size does not fit all. He also said that SQL by itself is not wrong but its implementation is. It is interesting to compare his remark with Scott Jarr's, who said the same thing. Nati predicted some sort of standards would emerge by consolidation but not from standards bodies.

After conducting five interviews, I have some idea of what NoSQL is all about. One thing I am certain of is that the utilities business is increasingly dependent on ICT technologies. Without them, smart grid will not be accomplished.

I sat down with Metric Insights' Marius Moscovici, CEO, and Steve Mock, COO, to find out what they were up to.

From left: Marius Moscovici, CEO, and Steve Mock, COO

When I interview companies at conferences, I usually want to know what they do in relation to the conference theme, and what their differentiation, future directions, and competition are. What they do seems to be fairly easy to understand, but differentiation and competition may not always be that easy to figure out. The NoSQL-related domain is still being defined as it moves in many directions at Big Data creation speed. Certainly, we need solid database technologies, including databases themselves and any utilities that enhance them, analytics engines, and good visualization tools.

For further expansion of this market, it is vital to get buy-in from the enterprise. William McKnight in his keynote speech advocated for putting NoSQL in the enterprise market and emphasized that only then would the NoSQL market become legitimate. Riding on what the enterprise already embraces would lower entry barriers for NoSQL-related technologies and services.

Business intelligence has been a big push in the enterprise market. Even before the age of Big Data, in a typical enterprise domain there was a large set of data not shared among different individuals and departments such as call centers, engineering, marketing, and HR. If a marketing campaign reflected call-center customer feedback, a company might be able to sell more of their products and services. Metric Insights' goal is to enhance BI capability by increasing each datum's value by associating it with more meaningful information. Because BI is already accepted in the enterprise segment, their goal is reasonable. They want to expand transitional BI into Big Data BI by sourcing data from Big Data as well as from traditional sources.

As shown below, Metric Insights collects data from multiple heterogeneous sources and adds context (relevant attributes, as shown in the figure) to make that data more effective and valuable. Metric Insights says this creates useful insights. The collected and context-enhanced data are stored in intermediate form (JSON) in a database. (By the way, some say that JSON will push XML out completely, and others say, not so fast, because the world is not built by Web alone. But that is not the focus of this blog.) When data have more attributes or context, you can provide more effective analytics because you have more relevant information on each datum.

Metric Insights’ system consists of data collection, augmenting data with other attributes (context), analysis, and visualization. By the way, I mentioned to those gentlemen that I was more interested in how the backend works than in their user interface (UI). Yes, the backend is important, but the front-end, the UI, is crucial in the BI segment. So gentlemen, I take back what I said. My comment came from my techie point of view. When you use a BI system, the first thing people pay attention to is the UI. Because not all BI users are data scientists, BI specialists, techies, geeks, or interested in how it works, its use should be easy and intuitive without lengthy training. When you present your BI tool, if it does not communicate its ease of use and simplicity, no one will pay any attention to it.

Metric Insights prepares typical dashboards for ease of use for a given application. The example below is for a sales database. Sales reps just select a pane to get to what they want instead of creating complex queries (like "what I would like to do”) to obtain the result they are looking for.

A typical screenshot of the UI for the sales database is given in the following. This example shows product releases and the number of daily sales demos made. When a new release is given, it is likely to have an increased number of daily demo requests. But if there is any sudden increase or decrease, you can take a look at that particular point and drill down because more relevant context is available and can be added as an annotation.

Architecturally, it uses a persistent cache to accommodate real and semi-real time data speed and store data in a local Mysql database as well as a document-based store (JSON format). Since it is document based, it is easy to add more information for each datum. Their system works with some well-known Big Data storage/databases and technologies, such as Cloudera, MongoDB, and Google BigQuery. Additionally, a secondary memory-based caching layer is used to optimize end-user access speed of analytics.

Their application areas include sales, production, inventory, and finance, and they are expanding their scope to include recruiting talent. This is an interesting area. It used to be difficult to gain information on each individual because publicly available personal information was very limited. A résumé is written to cast the best light on the job applicant, and references usually provide only positive comments. Now in the era of SNS, we can gather a vast amount of information on individuals when they are at their ease and off their guard, so to speak.

In a way, Metric Insights and Fluid Operations provide a similar product. They both collect data from multiple sources, convert them to a standard form with additional information, apply analytics, and visualize results. On the surface they are similar, but their focus and implementation differ significantly. Metric Insights uses context to enhance each datum and obtain insight, then stores it in JSON-based storage, which is more common for NoSQL players and more relaxed (and easier to manage) than the semantic model Virtual Operations uses.

I think both approaches are valid, and each has its good application areas. The market is still evolving and is big enough for both of them. I asked Metric Insights if they have considered the power industry as an application area. They have not considered it yet, but I think their product can be used for that. The power industry will face multiple Big Data problems, as they will have more real-time monitor data, such as meter-read, equipment status, data feeds from other systems like weather, static information like assets and service logs, and SNS. A utilities backoffice is filled with disjointed applications without much data sharing, which can be improved very much by something like this technology. I do not know how, but that is up to folks like Metric Insights and Fluid Operations.

Fluid Operations (FluidOps) aggregates data from different sources and converts them with some intelligence for better analysis. I sat with Peter Haase, senior architect, and chatted about their Information Workbench, a comprehensive tool for collecting and analyzing data and visualizing useful information.

Peter Haase

Fluid Operations is located in Walldorf, Germany. SAP's headquarters is there as well. They currently have no US office, but their website provides information in both German and English. Peter and other people from the company are fluent in English.

As in other areas, in the power business, utilities companies collect and aggregate various kinds of data in addition to meter-read data. They may monitor equipment on the distribution grid, such as transformers, switches, relays, and capacitor banks. The data from the equipment and the meter-read data may be generated at dramatically different speeds. In addition to dynamic and real-time data, some static data types like asset information, including equipment location, brand, model, specification, and service records, may be required to provide preventive maintenance and report malfunctions and failures. The FluidOps solution is to collect and aggregate data from multiple sources and then to translate each datum semantically to a common form so that it has more meaningful information associated with it. Since all the translated data are in the same form with more meaningful relationships among them, analytics becomes more effective and can lead to more appropriate action.

"Semantically” means that they convert collected data into their normal form, which is represented using the Resource Description Framework (RDF). I will not get into details here. Although it is not the same but in a way, it is similar to Entity-Relation model. An example diagram is shown here.

An example RDF graph (Source:Eric Miller)

All the data collected are converted into this format. The query language for RDF is SPARQL Protocol and RDF Query Language (SPARQL).

FluidOps Information Workbench consists of data integration and storage, data management, and presentation/interaction/UI customization layers. At the 30,000-foot view, it collects and associates data using semantic models from diverse industry segments. For example, the Linking Open Data Community project is an attempt to make data from different industry segments freely available, and for that, data are represented in RDF. The segments include media, geographic, publications, user-generated, governments, and life science. Their relationships are shown in the following diagram, which is maintained by Richard Cyganiak and Anja Jentzsch.

Linking Open Data cloud diagram, by Richard Cyganiak and Anja Jentzsch: available here. Data published in Linked Data format based on the Linking Open Data Community project.

Click each circle on the figure here (not the figure above) to drill down through each dataset.

The following figure illustrates how Information Workbench collects and associates data with other data to increase their value semantically.

The architecture of Information Workbench is shown below. It consists of a data integration and storage layer (green), data management (brown), and presentation, interaction, and UI customization (blue).

I talked about each utility's operation, but if we look at each region, such as ISO/RTO, the regional power balance information and data are very useful. I would like to follow this as it grows.

Bob Wiederhold, President and Chief Executive Officer, Couchbase

Couchbase was formed about a year and a half ago (February 2011) by merging Membase (based in Mountain View, CA) and Couchone (based in Oakland, CA). Bob came from the Membase side, and the new Couchbase is located in Mountain View. Couchone was behind Apache CouchDB, which is open source written in Erlang (also open source) and with Apache License 2.0. Most of the original key developers and committers (including Damien Katz) for Apache CouchDB moved from Couchone to Couchbase. The original developers and committers still contribute to Apache CouchDB, but most efforts are now focused on Couchbase 2.0, which is a separate open source project also licensed with Apache License 2.0 and is being implemented mostly in C. This is because Erlang is a functional programming language and C is more appropriate to increase speed. I’ve dealt with many programming languages in the past but never touched Erlang before. Bob emphasized that while Couchbase is heavily influenced by Apache CouchDB, it is a completely separate open source project. Bob told me that the merger went very smoothly and they are now about 100 people strong.

                             +

                      ||

I told Bob that I am confused by the NoSQL market, and he shared his view of it. It is interesting to hear different persons' views on the market. Of course, there is not 100% agreement on the current market, but different views sometimes give me a pretty good perspective. He first distinguished the operational from the analytics engine, as below. The analytics engine is a Hadoop and its derivatives, such as Cloudera, Hortonworks, and MapR. Note that Couchbase is a partner of Cloudera.

Then he expanded the NoSQL area according to technology and placed NoSQL players in each category. I will not discuss each category in detail. Those who want more detail can reference here. Wikipedia classifies the NoSQL categories in a much finer way. For example, there are several subcategories for the key-value camp, and it distinguishes the graph-based from the object-based ones. By the way, at the conference I took a four-hour crash course given by Dan McCreary of Kelley-McCreary & Associates. It was a good tutorial, and if you had the chance, you could sit down and spend a half day in his class. I also thought a whitepaper by Couchbase, Navigating the Transition From Relational to NoSQL Database Technology, useful. It describes document-based technology in comparison with the relational database.

The current version of Couchbase (1.8) is in the key-value camp. But come the 2.0 release, it will become a document-based database completely. Each camp has it merits and shortcomings. Will one category dominate others and all its technologies be consolidated into one? As for what will happen to this market, Bob thinks the following.

He thinks the key-value and the document-based databases will be merged, and the merged area will be the biggest of the three new areas. The other two areas will not go away but remain somewhat a niche market. The document-based solution is powerful, as it can contain a document like an entire website as a blob (in a JSON format) and retrieve it. For this, JSON is becoming the de facto standard over XML; Couchbase also uses JSON. There are proponents for both JSON and XML. In the Web environment, JSON is far more suitable, but XML has its own areas of application. There are a few tools for converting JSON to XML and vice versa.

As for the competition, Bob was very frank in analyzing Couchbase against other players in the document camp, as in the following table. Checkmark size indicates how strong and complete an attribute is. Well, the size is somewhat arbitrary and just indicates relative competency. Bob said that Couchbase has put a lot of emphasis on performance, scalability, and always-on features (thus, big checkmarks) with less focus on ease of development (thus, a smaller checkmark). He also added that with the 2.0 release, ease of development will improve significantly since this is the point at which they become a document database. He said that his competition has put a lot of emphasis on ease of development but needs to work on other features.

                                         Couchbase moves to focus on ease of development

                                          Competition moves to other features

He said although ease of development requires a lot of expertise, other things, like performance, are very hard to improve. He told me Couchbase has a big advantage in that it can consistently provide sub millisecond latencies for reads and writes that is often 1/3 to 1/10 the latencies of other solutions. In addition, Couchbase can provide throughput/server that is often 2-4x higher than competing solutions (see http://bit.ly/NKJkVH and http://bit.ly/Qulb4R). The consistent low latency assures very responsive applications and the higher throughput per server means you need to buy less hardware and software than with other competing solutions.

The current application areas that use Couchbase include social gaming, ad and offer targeting, social networking, online business services, e-commerce, cloud data services, and mobile-to-cloud data synchronization. Because I am interested in the application of NoSQL technologies to power utilities companies for smart meters and monitoring (such as with sensors with SCADA access) many types and speeds of data (static, like asset data, to real-time meter-read data), I wondered how products like Couchbase can be applied. Bob’s view was that as

the amount of sensor data and the frequency at which it is gathered increases, having a central database that can keep up with the inflow of data will become a challenge. NoSQL databases that have an ability to linearly scale up write throughput are an easy solution to capture the incoming data stream. Techniques like Couchbase Server's incremental map reduce are ideal to provide real-time aggregation/analytics over the data.

I asked him about an ecosystem for each player. He thinks developing an ecosystem is vital for the success of Couchbase. The way things are, the market seems still very confused, but it is expanding rapidly.

Scott Jarr

When I was researching the NoSQL segment, I found it confusing enough, but there is also a NewSQL movement, which confused me further. The NoSQL movement began in an effort to accommodate the Big Data phenomenon. In the traditional database segment, ACID—atomic, consistent, isolated, and durable—is of utmost importance. The relational database was developed to guarantee ACID and for transaction-oriented applications. The traditional relational or SQL database is fine, as long as the data comes in at a reasonable speed and volume and is of limited variety. But at some point, these parameters exceeded what the traditional SQL database could handle, and new ways to cope with them were increasingly required. That is where NoSQL comes in. NoSQL, in general, does relax some of the rigid SQL rules (abandoning SQL partially or altogether and thus ACID) and accommodate these new requirements; i.e., scale-out, high availability (HA), replication, and performance. Therefore, NoSQL in general does not have SQL, relational schema, joins, or ACID (this is obvious since these are traits of the relational/SQL database). Scott put the comparison of Old SQL, NoSQL, and NewSQL on a piece of paper as we spoke. I reproduced it here. Old SQL (yet another term) refers to the traditional relational/SQL database that dominates the enterprise world.

Comparison of Old SQL and NoSQL

So in other words, in order to gain scale-out, HA, replication, and performance, NoSQL abandoned SQL/relational schema partially or altogether. What NewSQL is saying is that it can accomplish every feature in the table above while keeping the relational/SQL schema (and therefore ACID and join).

If that table is expanded with NewSQL, we have the following.

Comparison of Old SQL, NoSQL, and NewSQL

How can that be possible with NewSQL? Performance gain is a result of new architectures that remove the old baggage of OldSQL, many leverage memory for additional improvements. Actually, Gigaspaces, which I also interviewed and will write a blog about later, has an in-memory cache technology working with other NoSQL companies. However, Michael Stonebraker, CTO of VoltDB, said in one of his talks (available in a 30-minute video) that running in memory alone does not guarantee the performance gain needed to accommodate the speed at which Big Data comes in.

Mike explained in his talk that there is nothing wrong with the concept of SQL itself. It is the implementation of SQL that causes the problems shown in the table. Because of the less than perfect implementation, 96% of the time is spent on overhead and only 4% on useful work, as indicated in the following graph extracted from his presentation.

   CPU cycle use in a typical SQL implementation. Most of it—96%—is used for overhead.

Unless these overheads are removed, even if all data is placed into memory, extreme performance improvement is not expected, because it only addresses the 4% but not the rest of the 96%. Typical NoSQL databases abandoned or partially supported SQL to bypass this problem. VoltDB faced the current inefficient implementation of SQL and developed their version of the SQL database from the ground up to eliminate these overheads. I am not covering each overhead in detail, but you can watch his easy-to-follow video.

OK, I get it. Then, what does this mean to the whole area of NoSQL? Does this mean the whole area of NoSQL gets consolidated into a single technology like NewSQL? Scott drew me a good figure to explain this, which he had already published in his own blog posts (the figure below came from part-1 and part-2).

From Scott Jarr's blog. On the Y axis, data speed, size, and complexity grow upwards.

There are five areas to address in the enterprise in terms of data collection and analysis (analytics): interactive, real-time analytics, record lookup, historical analytics, and exploratory analytics.

The five areas are further explained in the following figure with applications and time scale.

Note that VoltDB is colored differently from the rest of NewSQL in the graph, but that is meant to emphasize its position in the NewSQL group. VoltDB falls into the NewSQL camp. Scott emphasized its performance superiority over others. The performance benchmark they share is 3 million transactions per second (TPS). According to Scott, the traditional RDBMS is trying to cope with the Big Data problem (velocity, volume, and variety) by scaling up (throwing in more CPU and storage power rather than using parallel computing).

In Stonebraker's video, he said that VoltDB was five times faster than Cassandra and also faster (he did not say how much) than an unnamed incumbent's database. When I consulted with MySQL, it was very fast (before their 5.0, which incorporated enterprise-ready features) and faster than this incumbent's, but they could not publish the benchmark for fear of a lawsuit. I can understand that. When I consulted for JBoss, a Japanese open-source consortium compared their performance with other products like IBM's Websphere, without any tuning. The number was not very good, mainly because those who ran the benchmarks did not know how to tune JBoss’s compared to IBM's. After a JBoss engineer flew over there and tuned it, it improved drastically. So when we conduct a performance comparison, we need to set up a ground rule for comparison for every participant.

Of course, there are some overlaps among those technologies and their areas of applications, but this figure is a good picture of how each technology is suited for its application area. Hadoop is batch processing based and is not suitable for real-time analysis. Many people think Big Data and Hadoop are synonymous, but this clearly shows they are related but not the same. In the utilities business, a large amount of meter-read data gets collected, aggregated, and stored. By daily or monthly analysis of power usage for a particular area, a utilities company can probe into usage patterns and trends. Actually, some utilities are using Hadoop now, according to the Soft Grid conference.

Scott thinks NewSQL, NoSQL, DataWarehouse, and Hadoop will remain separate technologies because each of them is suited for some specific area of data collection and analytics. But he advocated that these areas and their tools be tightly integrated to provide analytics and thus effective real-time actions, as in the following figure. By incorporating the analytics results for long time spans into short-time analytics, more effective actions could be obtained.

Finally, he showed the current applications of VoltDB, as follows.

I consulted for MySQL before and knew something about the relational database market. But after it was bought by Sun, I stopped following it. I knew there was such a thing as NoSQL but initially thought it was "No to SQL”; it is more like "Not only SQL.” NoSQL started to get attention circa 2009, and the NoSQL Now conference was only started in 2011. So it is a relatively new area and, as in any new area, the market is very confused. Many terminologies and acronyms are floating around, with many claims by vendors. Quite frankly, it is very, very hard to walk through this market without getting totally confused. Prior to attending the conference, I studied the companies I planned to interview and read anything and everything I could put my eyes on. The sad reality was that I was further confused.

What is NoSQL, technology-wise, component-wise, and application-wise?

Technology-wise

The NoSQL market can be described in a few ways. One way is to categorize it by the technologies used. The 451 Group's Matt Asllet, in his blog NoSQL, NewSQL and Beyond: The answer to SPRAINed relational databases, gave a pretty good picture of the market, with categories and the vendors who belong to each category.

Matt Asllet's database categories

This figure alone is very valuable. This figure helped me to understand where my interviewees’ companies fall in.

This view is great, but I was still not comfortable enough to say, "Yes, I got it.” Bob Wiederhold, president and CEO of Couchbase, made it much simpler for me.

Bob Wiederhold

He thinks NoSQL is playing in a segment that is not concentrated on by NoSQL players that are good at transactions or suitable for backoffice applications. He further classified NoSQL into four categories:

• Key value

• Document

• Column family

• Graph

The current Couchbase (1.8) belongs to the key-value camp but will move to the document camp at its 2.0 version launch. He also told me that the key-value and document camps are being merged and the combined camp will be the biggest of the three new categories. I plan to write about his interview in a future blog.

How they fit together in the enterprise

How do NoSQL technologies fit into the enterprise? William McKnight, of McKnight Consulting Group, presented a keynote speech titled, "Putting NoSQL in its Place—in the Enterprise.”

William McKnight

One of his slides shows really well how data is collected, aggregated, and analyzed in the enterprise, and which components are there for each function. Data are collected for analysis; otherwise, there is no reason to collect them. There are two major groups for analysis: real time (streaming) and static (stored data). In his slide, Hadoop (which processes data in batch mode) is placed on the analytic side. But if you need to analyze a massive amount of data as it comes in real time, you need streaming analysis. Hadoop is not meant for that. That is why we need databases that can handle real-time streaming data, which is in a totally different area from that of Hadoop.

In the picture, blurry brown lines indicate a set of clouds. The components surrounded by the brown lines may be hosted in a cloud.

Application Areas

This is great. Then, what about application areas? Where does each NoSQL technology apply? Scott Jarr, cofounder and chief strategy officer at VoltDB, gave me the following figure.

Scott Jarr

Actually, he drew this on a piece of paper but he had a published blog. I will cover it in more detail in a future blog. He looked at the five areas of applications: interactive, real-time analytics, record lookup, historical analysis, and exploratory. He then placed each Big Data technology in one of the five areas. This is a pretty good explanation of NoSQL in terms of application areas. In the figure, VoltDB is colored differently from NewSQL, but he classified it in the NewSQL camp.

Applications to Energy (Smart Grid)

The applications areas discussed most throughout the conference were publication, financial, and SNS. A couple of people said that SNS is a driving force for Big Data and NoSQL on the West Coast, but on the East Coast it is primarily financial communities. What about its application to smart grid? In the Soft Grid conference, focus was on metered data, which will be collected, aggregated, and stored in real time but analyzed in no real-time fashion. I heard during the Soft Grid conference that some utilities were using Hadoop to analyze their metered data.

The Northeast blackout of 2003 was caused because timely actions were not taken to isolate the problem area from the rest of the power grid, and faults cascaded to the entire area. The causes of the blackout were studied intensely. But in 2011, it was repeated in the San Diego area. The initial cause may be different from the one in 2003, but the impact cascaded in the same way as in 2003. With the more connected ICT technologies, modern monitoring systems like SCADA, and real-time analytics of power grid health, this could be avoided. The decision to cut off faulty areas from the grid requires real-time action by monitored data coming in in real time because power moves very quickly. This is an application area that is different from the trend analysis done with Hadoop.

Those companies I interviewed told me the application to smart grid may be an interesting idea, but it is still premature, as they do not see the market forming. Finally, I just want to mention that David Brown of EMC, a parent company of VMware, used GemFire to implement data collection and analytics for some unnamed utilities. His case was an exception, and I guess the market is still being formed for the utilities.

David Brown

Data Center Infrastructure Management (DCIM) has been given a lot of attention in the past few years. I think we are moving from "What is DCIM?” to "How do we deploy it?” DCIM provides a lot of different functions, covering a wide range of areas in designing and operating data centers. There is no agreement on terminology or standards yet, but the functions may include inventory, change control, capacity management, simulation, and monitoring/reporting. And maybe more.

Let me take a different view here. Yes, functions are important, but functions require data available from a data center. In general, DCIM collects data associated with the infrastructure, i.e., power consumption, temperature, pressure, humidity, fan speed, and other data. What's missing is logical and virtual data. Although facilities equipment for HVAC and power delivery have both physical and logical parts, they are basically very similar in nature. IT equipment, like servers, has two distinct parts: physical (hardware) and logical (software). DCIM has taken both facilities and IT equipment in designing and operating a data center. Moving ahead, we may want to consider how we can make DCIM even more effective. This is shown in the following figure.

Before virtualization and cloud computing, each application had a dedicated physical server and did not move away from the designated server, regardless of its utilization. As each application is virtualized and tied to a virtual server in the form of virtual machine (VM) and multiple VMs are placed on a single physical server, each VM may be created, killed, or moved from one physical server to another in real time. This is done to reflect application utilization. If an application is not used much, it may be stopped or killed. If enough VMs are removed from one physical server, it may no longer be necessary to run that physical server, and it can be shut down or put to sleep. If the physical server's utilization is down, so is the cooling and power requirement for it. Facilities equipment for this may need dynamic adjustments in real time. So we need more tight integration of the virtual and physical sides of a data center for effective operation.

I will be moderating a panel on this issue, brought by BrightTalk at 9 a.m. (PDT) on August 16,

DCIM: Managing the Facilities and IT of the Data Center - Panel Session

We have a terrific set of panelists, listed below. Please join me in the discussion.

Moderators:

· Peter Gilbert, VP, Energy & Sustainability Solutions Strategy, CA Technologies

· Derek Schwartz, Deputy, IT Operations & Maintenance NIEHS at SRA International & Executive Director and Founder of The Green Data Center Alliance

· Robert Haggerty, Mission Critical Operations

Google has been in the PaaS space but recently dived into IaaS. A great thing about living in Silicon Valley is that local meetings with the movers and shakers of technology market segments are within an easy 30-minute drive. The meeting that prompted this blog was on cloud computing. Cloud computing is changing the way IT is delivered and thereby changing how and where we conduct computing. Consequently, it has a vast impact on the way we use energy. Some conclude cloud computing saves energy, while others are skeptical about it. In any event, Google, a mammoth of computing power, has entered another category of cloud computing, so I attended SVforum's cloud computing and virtualization special interest group meeting on Google's new service. (By the way, Google Compute Engine was announced about a month ago at the Google IO conference.) The SIG meeting was well attended; more than 100 people showed up.

These were the speakers:

• Marc Cohen and Kathryn Hurley of Google, developer relation engineers

• M. C. Srivas of MapR, cofounder and CTO

Dave Nielsen, one of the cochairs of this SIG, and others did a good job organizing this meeting. Normally, when a new product or service is introduced, a vendor or provider gives a dry presentation about how great their product or service is. Google wanted to show how solid their high speed and security offering is, so they included a couple of good demos.

Well, that was not enough. On top of that, they invited MapR, one of their partners, to share their experience in deploying MapR's version of Hadoop on the Google Compute Engine platform. Hadoop is open source from the Apache foundation and is used to crunch Big Data. MapR took Hadoop and added features to make it compatible with enterprise requirements.

Big Data and Hadoop have opened up a tremendous number of market opportunities, so MapR is not the only one to exploit them; Cloudera and Hortonnetworks also provide Hadoop on steroid versions for enterprises. By the way, I found a research report by Dave Ohara, The big machine: creating value out of machine-driven bigdata, to be very good tutorial material on Big Data.

MapR also has deployed its engine on Amazon AWS, and a performance comparison would be interesting. During the discussion, MapR revealed their performance data but said they had not conducted such a performance benchmark on AWS.

Back to Google Compute Engine. Google is no stranger to cloud computing. Here's a simplified list of Google's cloud offerings:

Marc Cohen speaks about Google's history of cloud offerings.

What is Google Compute Engine? Marc summarized it on one slide:

Marc Cohen presented an overview of Google Compute Engine.

In short:

• Infrastructure as a service (IaaS)

• Supports Ubuntu and CentOS (more operating systems, such as Windows, will come later)

• KVM as hypervisor

• Deployable in two territories (one eastern and two central time zone data centers, only in the same data center rather than inter–data centers)

• In private beta (need to apply to be deployed on the platform; see here for more detail)

• Free for now (only qualified users) but will be charged for later

• No SLA guarantee now (under consideration when released officially)

Marc did not elaborate on SLA, but judging from what he said, you need to specify which territory you want to deploy your load in. All the computing and data associated with it stays in the same data center (i.e., cloud). No matter how we improve technologies, we will still be bounded by the laws of physics, and we cannot send packets any faster than the speed of light. If they want to guarantee SLA, they need to make a lot of assumptions and impose restrictions on their customers. I have not heard any cloud service provider discuss SLA, and I wonder how they can guarantee it, even with conditions.

More details can be found in their data sheet here, and here are two useful links:

• Google's Cloud platform

• Google Compute Enginer Documentation

Some architectural information follows.

The yellow disklike box behind Kathryn Hurley is cloud storage.

One thing unique about this is the use of their command language interface with gcutil library.

Now to benchmarks and usability shared by MapR. It is more convincing if your actual user, rather than yourself, says good things about your offering. Even though this is not a blog on Hadoop or MapR, here's some basic information shared by Jack Norris, VP marketing, of MapR.

MapR's summary:

He also summarized MapR's deployment on Google Compute Engine, as follows:

Srivas added the following, more-detailed benchmark:

The one on the Google platform outperformed that hosted on a physical platform in every benchmark except for processing time:

There was no comparison in cost. Srivas said jokingly that we had better think twice before purchasing and owning a bunch of servers. This is because a server's life is probably only two to three years, and the minute you buy a new server, it becomes obsolete because new servers with new technologies are invented constantly. Spending so much money is one thing, but after your business goal changes and you no longer need that many servers, what do you do with them? They do not disappear magically.

In the area of IaaS, AWS was way ahead of the market curve, followed by RackSpace and others. As Dave Nielson said at the beginning of the meeting, those who were working on IaaS (e.g., Amazon) are adding a PaaS solution (e.g., Elastic Beanstalk) and those who were in the PaaS market (e.g., Google) are adding an IaaS solution. The cloud market is still expanding, and in spite of some problems, such as lock-in (due to no standards), security worries, and lack of control, because of its wide and broad market, it is still growing rapidly. As long as mobile computing and sensor networks (such as smart grid) are growing, it does not seem that the end of growth is even near. When very few interoperability cloud platforms exist, it is always good to have competition.

Because a lot is going on with IT, this panel was very interesting. As more and more people and devices and equipment (without human intervention) get online, the importance of data centers increases even more.

First, here is a list of panel participants:

Moderator

• Bruce Myatt, M+W Group

Bruce Myatt

Panelists

• Peter Gross, Bloom Energy

• Ron Hughes, State of California

• Zahl Limbuwala, BCS Chartered Institute of IT

• Ben Radhakrishnan, Lawrence Berkeley National Laboratory

From left: Peter Gross, Ron Hughes, Zahl Limbuwala, and Ben Radhakrishnan

The following is a summary of the discussion, with my comments (indicated by ZK). Bruce started by saying how, since the 2007 EPA report on data center energy consumption, we came to increase data center energy efficiency by means of several technologies and methods. Those include air economizer (free cooling), cold and hot aisle separation, cold or hot aisle containment, and measuring, which led to the wide deployment of the PUE metric and more energy efficient IT equipment. Other than more energy efficient IT equipment, these are primary from the data center operator’s view. Then, Bruce posed two questions.

Question 1: Types of changes in IT operations in the next 5 to 10 years

The first question was the type of changes taking place now and what will happen in 5 to 10 years that will have a big impact on IT operations.

A number of such changes were discussed, such as:

1. Emergence of mobile devices (specifically tablets, whose shipment will reach 400 million in 2016)

2. Cloud computing/virtualization

3. More security needs

4. Much bigger scaling changes for each server refresh

5. Emergence of ultra–energy efficient HPC gear for supercomputers that encourages cloud computing operations

6. Change of emphasis from facilities to IT

7. The CIO’s role will change to that of a broker of services who decides where each application resides, either in-house or hosted, from that of a provider of infrastructure services.

(ZK: Most points raised were not accompanied by a detailed explanation, and I would like to express my own opinion about their impacts on data centers. The second question below was supposed to answer that. However, there was no clear one-to-one correspondence between the questions in question 1 and the answers given in the discussion of the second question. That does not mean the discussion was not useful, though. For reference purposes, I’ll refer to each question as 1-x, with x being the number associated with it.

Actually, 1-1 through 1-5 are very much related. Although virtualization is not cloud computing, it constitutes an important component of it. In short, virtualization with factors like policy and automation would create a cloud. As we have more mobile devices like smartphones and tablets, it becomes necessary to take care of them at a data center. Because the scaling factor would be several orders of magnitude bigger, because of more new IT equipment processing power and the sheer number of client devices, it will become necessary to have an infrastructure that can handle such a scale. That is cloud computing. As computing is concentrated in clouds, more security needs will be emphasized. Security is not limited to perimeter security of a data center. Security can be broken within the data center and via networking as well. Hacking and malware will become more sophisticated, and when security is bleached, its consequence would be several orders of magnitude greater than when everything was processed manually on paper.

As to 1-6, I also felt that more emphasis was placed on the facilities side in data center conferences before. But conferences, media, and blogging are now beginning to pay more attention to IT for improved data center operations and energy efficiency. This is very important because, after all, facilities exist to support the reliable and stable operations of ICT equipment. In turn, ICT equipment is run to support the business goals of an enterprise. It is easier to see concrete numbers like kilowatt hours consumed and/or temperature and humidity than to see what applications are doing for services.

As for 1-7, cloud computing with automation will relieve a CIO and his staff from repetitious and routine tasks, and they will become knowledge workers. See my previous blogs on this subject here and here.

As cloud computing moves into the mainstream of computing, some operators will abandon their own data centers and move their computing to cloud providers. Although data centers will not disappear, the number of data centers will decrease. At the same time, cloud providers (cloud data center operators) will expand their current data centers at their current and new locations. Each cloud data center in the future may be bigger than current ones, requiring more power and cooling. Or future data centers may require less space and power (thus, less cooling energy) because of new technologies like supercomputers with an enormous amount of processing power and very high energy efficiency.

Question 2: How those changes with IT will affect the design and operation of data centers

The second question was to solicit opinions on what those changes would bring to data centers.

(ZK: As I said, there was very little one-to-one correspondence between the first question and the second question. That is why I injected my own opinion above.) These were some of the points raised during discussion of the second question:

1. A method of assessing data centers from environmental, economic, and specifically benefit points of view will be developed.

2. One proprietary system providing IT functions (hardware, software, server, storage, and network in a bundle) will be used more.

3. Self-management, including self-healing, in the IT segment will emerge for ICT equipment.

4. Cloud computing will become more prevalent than data centers.

(ZK: I refer to each question as before.)

As for 2-1, a fair and thorough assessment method of data centers will be developed. A data center should be evaluated from environmental, economic, and societal benefit points of view. Currently, the big emphasis is on economics.

(ZK: Often data centers receive beatings from environmentalists because of large space, power, and water consumption. But there should be a fair and comprehensive assessment because a large data center may be more energy efficient than small data centers and provide essential societal benefits.)

(ZK: Regarding 2-2, an embedded solution is generally more efficient than an open solution. The ICT segment has made significant progress by providing open solutions. But when it becomes really necessary to focus on energy efficiency, it may make sense to develop a system that combines hardware, software, server, storage, and networking together, such as Cisco's unified computing. Within the system, some proprietary protocol may be used, but if IP is used to interact with the system from outside, there should not be a big problem.)

As to 2-3, servers and other ICT equipment will become more self-managing and self-healing. That will cut down human intervention in case of minor to medium-size problems.

(ZK: This may be implemented with redundancy and by moving virtual machines from a faulty server to another to continue processing. Cloud computing does some of this already, but more progress will be expected.)

Regarding 2-4, I already discussed this above. But this opinion was expressed by Ron Hughes of the State of California. His actual comment was that state and local governments will no longer build new data centers but use clouds. (ZK: The US government has indicated this as well, and it also applies to the private sector.)

(ZK: As an IT guy, I welcome the shift in focus from facilities to IT. However, we need to move farther to incorporate data, such as server health and application status, from the higher layers of ICT equipment. They are currently monitored by system management tools. Information like where each application runs and what status each application is in would have an impact on data center operations. More needs to be done in this area.

ASHRAE, founded in 1894, is a building technology society with more than 50,000 members worldwide. The Society and its members focus on building systems, energy efficiency, indoor air quality and sustainability within the industry. Through research, standards writing, publishing and continuing education, ASHRAE shapes tomorrow’s built environment today.

It has a big impact on data center operations. In the recent conference given by DatacenterDynamics in San Francisco, I had a chance to chair a track that included thermal guidelines from ASHRAE. Judging from the past attendance at ASHRAE guideline sessions, I was expecting a large turnout, and the two sessions on the subject were packed.

The first presentation was by Don Beaty, ASHRAE Fellow and president of DLA Associates.

Don Beaty

He was the cofounder and first chair of ASHRAE Technical Committee TC 9.9 (Mission Critical Facilities, Technology Spaces and Electronic Equipment). TC 9.9 is very relevant to data center operations because it sets guidelines for operating data centers, including facilities and ICT equipment. When I attend a conference session, I usually record it for accuracy and memory’s sake. But it was hard to do so as a chair. So I am recalling from memory and some of the details are a little bit fuzzy.

One thing Don kept emphasizing during the talk was that it is the temperature of inlet airflow to a server that matters for data center cooling but not the temperature in the room. In the past, CRAC units on the data center floor checked the temperature of returned air and used it to approximate the temperature of inlet airflow to a server. Obviously, it usually did not reflect the actual inlet airflow temperature. If cooling is via a raised floor, the inlet airflow temperature varies widely, depending upon the proximity of CRAC units. So it is imperative to measure and monitor the temperature at the inlet of each server/rack. At some point, cooling via a raised floor may not function well. For example, a rack that consumes 10 kW may not be cooled effectively with raised floor cooling. Furthermore, even though it is desirable to have uniform power consumption and heat dissipation from each rack, because of ICT equipment configuration requirements and other constraints it is not always possible to do so. Don presented a guideline for the inlet temperature for servers titled, 2011 Thermal Environments – Expanded Data Center Classes and Usage Guidance , and I extracted the table and a graph from page 8 and page 9 of the document, respectively, for reference purposes.

A psychrometric chart describes temperature and humidity and is used to set a proper range for the combination of the two in a data center. This chart shows A1 through A4 ranges, along with the recommended envelope.

The current server can operate fine (with server vendor warranty) with the A2 guideline shown above. A2 sets the high temperature at 35°C (95°F). But new guidelines can expand the acceptable ranges to 40°C (104°F) by A3 and 45°C (113°F) by A4. If you allowed this widely expanded range, almost any data center in the world could take advantage of free cooling, such as airside economizer. Incidentally, Christian Belady of Microsoft has said that developing a server that tolerates higher temperatures might raise the production cost (thus the purchase price), but millions of cooling dollars could be saved with several thousands dollars more for this type of IT equipment.

So what's holding up the production of servers with A3 and A4 guidelines? Next up were Dr. Tahir Cader, Distinguished Technologist, Industry Standard Servers (ISS), and David Chetham-Strode, Worldwide Product Manager, both of Hewlett-Packard. They shared very interesting results. Tahir is on the board of directors for The Green Grid, a member of ASHRAE TC 9.9, and a liaison between ASHRAE and The Green Grid.

Dr. Tahir Cader

Again, I do not have their presentations and unfortunately cannot refer to specific data. They experimented with power consumption at various geographical locations, using the A1 through A4 guidelines. According to their findings, you may not need to employ A3 or A4, depending upon your location. In many cases, there was little or no difference between A3 and A4. Sometimes there is some savings between A2 and A3, but it depends upon the geographical location.

When we consider the temperature in a data center, we also need to consider it for humans. Even though the primary purpose of the data center is to host ICT equipment and the temperature could be raised up to 45°C at the server inlet, doing so could also raise the temperature at other locations. Anything above 30°C may not be very comfortable for people to work in for a long time.

It was relatively easy in the past to pick your server, using some well-defined data, such as CPU speed, number of cores in the CPU, memory size, disk capacity, and number of networking ports and their speed. Even if you have data centers at locations throughout the world, you may buy only a few server types and get a good discount from particular vendors for all of your data centers. But another factor may be added when you refresh your servers next time, which is the analysis of the inlet airflow temperature to a server vs. power consumption. If you are big and sophisticated like HP, you could run your own analysis to decide which server (that supports A1 through A4) to use. But this analysis seems to be fairly complex and it may not be that easy. Being a chair, I needed to control the Q&A session but had a chance to ask a simple question. Can a server vendor like HP provide a service to pick the right type of servers for your geography? The answer was yes. That is good to know.

There will be five tracks or halls on different themes. I will chair Hall 2, whose focus is IT. Of course, the discussions will be on IT in conjunction with data center mechanical and electrical. Subjects at data center conferences tend to be on the facilities side, and very few are from the IT perspective. So I am very excited about this track.

In Hall 2, several subjects that relate to IT equipment and IT applications for data center optimization will be discussed. These are ASHRAE guidelines for telecom and IT equipment, cooling energy reduction, DCIM, and cloud computing.

Don Beaty of DLA Associates will discuss two guidelines from ASHRAE: 2nd edition of Datacom Equipment Power Trends & Cooling Applications and the upcoming 3rd edition of Thermal Guidelines for Data Processing Environments. This may lead to chiller-less data centers. Then Tahir Cader of HP will update us on the new server inlet air temperature guidelines known as A3 and A4. By raising inlet air temperature, we can cut cooling energy substantially. Are server vendors ready for the challenge?

John Boggs of Emerson will talk about how to reduce cooling energy at data centers with little or no cost to you. As we know, cooling can use close to 60% of power consumed at data centers. If we can control and reduce it (and with little or no cost), it would be great. I cannot wait for this presentation.

DCIM covers a broad area of the data center, ranging from design, simulation, monitoring, and controlling to retrofitting. The four DCIM sessions will collectively give us good insights into what DCIM is and what it can do for us. Two of the four sessions do not use the term DCIM in their titles or synopsis, though. The first one will be by Dhesikan Ananchaperumal of CA Technologies. He will talk about the necessity for good integration of data from both facilities and IT for optimizing data center operations. Jim Kennedy from RagingWire will share his experience with making both existing and new data centers energy efficient with DCIM tools, which led to EPA's Energy Star for data center certification. Following that, Khaled Nassoura of Raritan will tell us how to improve data center operations by combining DCIM with intelligent and automated systems. Finally, Todd Goldman of Nlyte Software will explain how to apply DCIM tools in steps without going too fast.

The past few years have brought an increase in cloud computing sessions at data center conferences like this one. Gina Tomlinson of the City and County of San Francisco will talk about how she put mission-critical IT infrastructures into a cloud. Although the notion of cloud computing is widely known by now, many people are still hesitant to move their mission-critical infrastructure into cloud for reasons of security, controllability, and SLA. How did she cope with the fear and move them to a cloud? It will be very interesting to hear about her experience.

Overall, this conference is filled with many more interesting sessions and speakers. But if you are an IT professional, come and join me for those very informative talks. See you in San Francisco!

When I first heard the term private cloud, I did not get it. You do not outsource computing or storage resources but use your own. Even if you do not use some server or storage, you still keep that server and storage in your building, and that costs you money. Something is missing here. Fundamentally, public and private are different beasts, although they may use the same technologies. In a way, with the emergence of private cloud and in contrast with public cloud, the notion of cloud computing has become more crystal clear.

Let's look at this from a different point of view. When you look back at the progression of your data center in terms of energy efficiency, you used to have one application for one dedicated server to make sure that application runs reliably and securely. Then came virtualization. Virtualizing your applications enabled you to host multiple applications on a single server, consolidating multiple servers to one for higher server utilization. Before virtualization, the server utilization percentage tended to be in the low teens, and multiple servers were required, raising power usage and costs. The next phase is automation of control. A bunch of virtualized servers do not form a cloud, but automation does. With proper management and automation, your virtualized data center becomes a cloud data center.

While I was thinking about this, Greg Ness, Chief Marketing Officer, Vantage Data Centers, sent me a link of a panel he moderated on this very topic. His session is listed at the bottom of the page. The video runs a little less than 30 minutes. Usually, you cannot do much in 30 minutes, but Greg covered several topics, including the definition of a cloud, public vs. private clouds, security, economic issues, and the future direction of IT as it relates to clouds. When you pack so much into a short session, the audience usually loses sight of what’s going and does not get much out of it. But Greg did a good job and moderated the panel to express very useful information.

The following were on that panel:

Moderator:

• Greg Ness, VP Marketing, Vantage Data Centers

Panelists:

• James Barrese, CTO, PayPal

• Winston Damarillo, Cofounder and CEO, Morphlabs

• David Nelson, Chief Strategist, Cloud Computing, Boeing

• Paul Strong, CTO, Global Field and Customer Initiatives, VMware

• Don Pickering, CEO, OneOcean

I will summarize the panel discussion with my comments below. If you want, you can watch the video here. Greg’s session is listed at the bottom. Unless indicated specifically by "ZK,” the following opinions are the panel's.

Greg started the session by quoting Gartner’s observation that cloud computing is now moving from the height of inflated expectations into the trough of disillusionment in its hype cycle. (ZK: Gartner's analysis was published in late 2010. The hype on cloud computing started with public cloud and software as a service (SaaS). In a way, public cloud is IaaS [computing first followed by storage]. Then came private clouds and platform as a service (PaaS). In 2010, neither private clouds nor PaaS were given much attention, though.)

The definition of clouds: The first topic was the definition of cloud computing. (ZK: In many panel sessions on cloud computing in the past, the discussion ended here.) Cloud computing features include fast provisioning, pay-as-you-go, and instant scale up/down. The consensus of the panel is that cloud computing is not a new technology but a new business model for the delivery of IT services. (ZK: No disagreements and so far so good.)

Public vs. private clouds: The next question was the difference between public and private clouds. Simply put, private clouds are an attempt to mimic public cloud technologies and the operating model, which are placed outside the enterprise, inside the enterprise. What private clouds could change includes faster provisioning and instant capacity adjustment. For example, the provisioning time was cut from, say, nine months, down to an instant. Once a new operating model is adopted and it works better than the previous one, no one wants to go back to the earlier model. The economic side plays a role in the adoption of private clouds as well. Add to this the process changes due to the new operating model, which makes both developers and guys in charge of architecture work better, and you understand why enterprises are embracing private clouds.

(ZK: In this segment, they did a good job of listing the merits of private clouds, and most people probably agree with their opinions.) Now the discussion got more interesting and covered why and how enterprises move their computing from public clouds to private clouds, as Zynga recently did. (ZK: This is something new to me, and I found their opinion very interesting.) The bottom line is ROI and the utilization of the assets in question. If your utilization of the asset is high, you should own it. If not, you want to outsource it. A good analogy is the ownership of a car. Unlike city dwellers, most suburbians like me own their own car, mostly because they use it daily. However, city dwellers can use other means of transportation and may not use a car daily. So it makes sense for suburbians to own a car and for city dwellers to take a taxi when needed. This is very reasonable and easy to understand.

Unfortunately, our life is not that simple. How do we handle spiky load increases, as in the case of disaster recovery or training, which may not happen very often? That is why we need a hybrid approach that uses both public and private clouds. The panelists advised using private clouds for predictable loads because it is less expensive to do so and using public clouds for spiky and dynamic loads. (ZK: This sounds good but here's a problem. When people talk about a hybrid cloud, they talk as if virtual machines (VM) running on a private cloud can be easily and seamlessly moved to a public cloud. But as far as I know, it is not that simple. VMware solutions are dominant in the enterprise market, and Amazon's AWS solutions control the public cloud segment. There are a few variants of cloud file formats, and those used by VMware and Amazon AWS are not compatible. Unless the file formats are translated, moving VMs between public and private is mere theoretical talk.) That is why one of the panelists mentioned the need for cloud standardization.

Another point that was interesting was the simplification and specialization of computing patterns. In a typical large enterprise, there are something like 6,000 applications. Each application needs to be:

• sourced

• designed

• integrated

• deployed

• managed

• monitored

• archived

To do this massive task, for the past 25 to 30 years, 80–85% of the cost has been dedicated to software and those who manage software. Cloud changes this. For example, companies like eBay and PayPal have only a few patterns, which cuts the cost of maintaining them. When you do this, your opex primarily becomes energy, and the issue becomes where you run your loads and how you optimize the power/infrastructure ratio.

At this point, Greg interjected an interesting fact drawn from his discussions with area experts. When your power consumption is less than 500 kW, it makes sense to outsource your computing to a public cloud. But if your consumption goes over 500 kW, it becomes more economical to have your own infrastructures (i.e., private clouds) so that you can tune them for your computing needs.

Security: (ZK: In almost any cloud discussion, security is mentioned as the #1 inhibitor of the adoption of clouds.) Is cloud security a solved matter? One of the reasons some people prefer a private cloud is because it appears to be more secure than a public cloud. Cloud security really comes down to data security. Cloud security needs to address regulatory requirements according to geographic locations. But they are being addressed. Cloud needs to be more secure. It is like spreading peanut butter over a slice of bread. If you spread it, it gets thinner. If you concentrate your security effort on one spot (a cloud) by an expert that does this as their core business every day, rather than spreading it over a mass (many organizations whose business is not security), security works better on the concentrated one spot.

The future of cloud: Some vertical clouds with specializations like financial, health care, specific compliance requirements (e.g., aviation), and government data will emerge. Who will dominate the cloud market, big companies like Oracle or small startups? So many different patterns are required to satisfy the variety of users. And one company probably cannot accommodate such a diverse set of requirements. For that, open-source clouds, such as Openstack, need to be taken into consideration. Although Openstack may be only 80% completed, it supports multiple technologies, such as hypervisor, work processes, and patterns.

In the future, when an enterprise user indicates what their core business is, a service provider like VMware will provide everything that is not strategic to its business and let the company do their business at their core. PaaS will be a differentiator when both SaaS and IaaS become commodities with little differentiation. An ecosystem will be formed around PaaS with specific IPs.

In 5 to 10 years, a lot could happen, such as development of massive amounts of memory/storage, low-power servers (e.g., 10 W computing), software-defined networks, support for multiple devices and platforms, global software development collaboration via pipeline, and standardization. These will change the way software with PaaS is designed and deployed, for sure.

(ZK: Even if you have read this far, I still recommend watching the video because I omitted some points and paraphrased some nuances. The video has a lot of subjects that are current, and you can understand where cloud computing stands at this point. Moreover, I sometimes moderate panel sessions, and I think I learned a lot about how to do that from this panel.)

So when a CIO from a data center talks, I do not want to miss the chance to hear what he has to say. In May, Kevin Malik, CIO of IO , gave a presentation on IO’s DCIM system, IO.OS, at the Uptime Symposium.

Kevin Malik

It was a very interesting talk, and I went to their booth to see a demo. I was very much interested in what they were doing in terms of facilities and IT integration. See, coming from IT, I still look at a data center from the IT point of view, and when it comes to monitoring and control, it is still system management. Leading companies like HP (OpenView), IBM (Tivoli), CA (Unicenter), and BMC (Patrol) provide system management tools. System management tools deal with higher stacks of the computing system, such as server health (online status), applications, utilization ratio, and response time, by using IP as the dominant protocol. They do not usually report infrastructure data like power consumption, temperature, and fan speed. On the other hand, DCIM deals with the infrastructure data for both facilities and IT (via Intelligent Platform Management Interface). From the computing system's view, DCIM deals with the lower layers of the stack. The following figure depicts this.

I wanted to talk to Kevin because of his practical knowledge of DCIM and his background as a software person.

The following is an edited version of my conversation with Kevin.

Kevin is CIO of IO and general manager of IO labs. As CIO he does what other CIOs do in non–data center companies, i.e., back-office support, such as HR and financial departments. That is fine, but his work as general manager of the IO labs is more interesting. He provides software solutions to what IO does. More precisely, he, who is in charge of software, works closely with William Slessman, cofounder and CTO, who is responsible for the hardware side of the data center. Many experts advise having a single management structure to oversee both IT and facilities for optimized data center operations. So before I even asked about their effort to integrate IT and facilities to operate a data center more efficiently, I knew they were serious about it. Their first goal is to control power consumption because the rate of data center power consumption has been skyrocketing and should be controlled. The best way to do that is to combine hardware and software to provide a solution. After all, the ultimate goal of having a data center is to run IT equipment for business, and a data center needs to work with IT harmoniously to optimize its operation for the best IT and business results.

As for the operation side of the business, Kevin may not influence site selection per se, but he makes sure that networking is available at the selected sites and that those sites are connected with major network providers of the likes of Level 3 and Abovenet. Also, he makes sure that each and every modular component shipped is hooked up with the proper software to be appropriately installed for accurate monitoring and control.

Their DCIM tool is called IO.OS and is available to all of their colocation customers. The customers can plug into IO's infrastructures, and all the power usage and other relevant information, including temperature and PDUs, becomes available through a web interface. Customers have the option to alter their environment, such as the temperature, if they think they can raise it, leading to relaxed cooling requirements (thus, less money). This is done on a customer-by-customer basis. However, at this point, as a service provider, IO does not have a say about which IT equipment the customers purchase or how they configure their IT gear at the colocated space. IO wants to work with customers to optimize the IT equipment to work seamlessly with the facilities infrastructure. For example, by monitoring power requirements for two servers running side by side, they know whether the utilization of both servers is low although a large amount of power is allocated, and they can consolidate these two servers into one to relax the power requirement.

I touched on the issue of integration between DCIM and system management tools above. IO is working on the integration of the two types of tools at this time. IO has a very robust web services layer to integrate products. Additionally, IO.OS supports integration with VMware allowing seamless communication with the data center and the virtualized stack. Any change to a data center, such as opening a door (which may increase the temperature) is considered a stimulus and captured by software. It then takes the necessary action, such as moving some virtual machines from one server to another, to best balance the power usage resulting from the stimulus. Kevin mentioned that what is not being considered now in the optimization formula is applications. System management tools deal with applications. If applications are taken into consideration, the whole data center from top to bottom would be considered, leading to even more optimized operations. So Kevin agreed that some integration between the two types of tools would help to make a data center operate in an optimized way.

Finally, I was curious about what they do in regard to using renewable energies for their operation. I was expecting an answer like solar and/or wind. But Kevin’s answer was somewhat unexpected. Pumped hydro storage is one form of energy storage. During the night, when the cost of power is the lowest, water is pumped up to higher ground. And during the day, when the power cost is higher, water is released to generate power (hydropower). In IO's case, ice is used in place of pumped hydro storage for cooling data centers. (Web Hosting reports some details along with a picture of such ice .) During the night, when power is cheap, they produce a bunch of ice, and during the day they use them to cool data centers. This may not be an application of renewable energies, but the idea is interesting and certainly shifts power usage away from peak times.

After MySQL's acquisition, many people left. Some I met after that, but others were pretty much gone from my circle of contacts. I did not talk to Brian Aker much when he was director of architecture at MySQL. The last time I saw him was at MySQL's Tokyo office with Larry Stefonic, president of MySQL Japan. I remember that the three of us had lunch after walking far under intense sun to find seats because many restaurants were crowded at lunchtime.

Brian Aker

In any event, I knew Brian was doing fine after MySQL but did not know where he was until recently. At a recent cloud meet-up, I was glad to find him as one of the speakers. His talk was about something called Gearman. It is not German by the way. What is it? He said in one word, it is an effective queue. Other descriptions are here.

Those who do not know what Mechanical Turk is can find the explanation here.

 On its own website, it is described as:

Gearman provides a generic application framework to farm out work to other machines or processes that are better suited to do the work. It allows you to do work in parallel, to load balance processing, and to call functions between languages. It can be used in a variety of applications, from high-availability websites to the transport of database replication events. In other words, it is the nervous system for how distributed processing communicates.

He is a fellow at HP now and in charge of Gearman. I did not know what Gearman was or how well it was received by the developer community. Brian said it is being used by Wal-Mart, Disney, Yelp, Craigslist, Flickr, Instagram (which was recently acquired by Facebook) and others. He knows this because he receives bug reports from these organizations. In running an open source shop, you learn that many people download your package for the fun of it. Some may actually install and use it, while others are content that they downloaded it and do nothing with it. So you do not know who is actually using your package until you hear from the users.

My understanding is cursory because it was the first time I heard about Gearman. But the main philosophy is to separate UI and the backend (Gearman's term is worker), and let Gearman be in the middle to control traffic. In this way, UI does not have to know anything about backends or workers. In a way, this is similar to object-oriented programming, which separates an interface from its implementation.

Using one of his slides in the following, UI requests a resize function but does not know or care how it is done, where it is done, or who does it.

The leftmost UI wants to do "resize.” The Gearman (its head is a gear) in the middle traffic-controls the resize command to an appropriate worker (far right) that knows how to do "resize.” Gearman looks at the requirements and spins on as many appropriate workers as needed to process a job.

The Gearman stack looks like the following.

The usefulness of Gearman:

As you can see from its architecture, you can spin off any number of workers, depending on the size of a load. Because Gearman can do aggregation, a large job can be chopped into smaller pieces and given to each worker. Upon job completion, Gearman aggregates the results together and returns them to UI. This is like Map Reduce or Hadoop. No wonder it is used by many SNS-based applications.

To conclude, here's how to obtain more information about Gearman. By the way, I forgot to ask why it is named Gearman.

Some regard social networking systems (SNS) as timewasters, but many others, like me, take them seriously. I am not sure if the current shape and form of SNS will continue to exist for many more years, but certainly mass participation and sharing of a vast amount of information in an elastic fashion will continue in one shape or another for many years to come. This means we need infrastructures to handle such requirements. I will try not to be too analytical, but these are two of those requirements:

1. Instant (by the second) increase/decrease of traffic of massive proportions with dynamic fluctuation

2. A vast amount of structured and unstructured data processing

I am sure there are more, but let's just make things simple by considering only those two. The first states the necessity of having a powerful infrastructure that can process an unseen amount of information. Because of dynamic load shifting, a static infrastructure is not a good solution. In other words, we need an infrastructure to satisfy somewhat conflicting requirements. One solution is to oversubscribe each and every component in the infrastructure. However, most times you cannot accurately predict the level of demand, especially for SNS applications. Because you cannot prepare an infinite amount of resources, this solution would be very hard to implement. Also, as stated in the second point, a vast amount of multiple formatted and unformatted data must be processed. On top of that, those data are coming in mass in real time. How do you handle that?

In a recent cloud computing meet-up held at Microsoft’s Silicon Valley office, there were two talks on accommodating these requirements.

The first talk was by Jim Zimmerman, CTO of Thuzi, which is based in Tampa, Florida. According to their website,

Thuzi is focused on making social media count.

Jim Zimmerman

To that end, they provide infrastructures and tools to assist clients' SNS-based marketing campaigns, which might be hosted on Facebook or other sites. Their tools provide a solution to the above two requirements.

Jim said that the techie population in Tampa is much smaller than that of Silicon Valley, and when they hold a meet-up, they can do it at a local Panera Bread Restaurant. In contrast, a good-size auditorium at the Microsoft Research facility was full of techies. I must confess that I could follow most of his talk, but here and there I was lost because I have not touched code for some time. But the problems he articulated were clear. They have a solution for both business and consumer applications.

A set of problems in dealing with business applications.

The problem statement is an expanded version of what I described. For this, their solution is here.

Command query responsibility segregation (CQRS) is not a very familiar term. Jim elaborated it further in the following architecture slide. Basically, it is a complete separation of UI and backend. The same philosophy was discussed in the subsequent talk by Brian Aker.

The consumer application has its own problems, as follows

And their solution for that is as follows.

They also have an application that manipulates photos. Their earlier version sent everything to the server for processing, but it took too much time. In the current version, they process the photos on the client side and send the processed photos to the server. This is summarized in the following slide.

The computing and infrastructure requirements need very different thinking to accommodate SNS applications. Public cloud is an ideal platform to accommodate unknown bandwidth because you can increase your computing and other resources on-the-fly. If traffic subsides, you can decrease the resources accordingly. So for you, it is good to save energy, but how about the provider? In theory, a public cloud provider has many customers running various loads simultaneously. Some may increase and others may decrease traffic at a given moment. So in the end, everything balances out and energy is not wasted. Intuitively, that seems to be true. But it really depends upon many parameters, including each cloud site's configuration and customer mixes. I think the jury is still out on this.

I cannot generalize their architecture but it seems that for scalability support, UI and backend processing should be separated to increase processing power, regardless of how light or heavy traffic may be. In the cloud environment, UI and backends can be anywhere. They may not be stationary at one physical location, and that makes the whole design very complex. Brain Aker addressed that problem in his talk.

The last slide contains the relevant reference information.

This year, the symposium's theme was unlimited bandwidth. It is very hard to define what unlimited bandwidth means. The more we use bandwidth for communications, the more we want.

There were several topics relating to unlimited bandwidth, including network infrastructure, next-generation technologies and applications, smart grid, and transportation. One thing I really admire about the Fujitsu team headed by Kevin Krejci is that they get very good speakers who are right for the symposium theme. My expectations were not betrayed this year, either.

My backlogs of writing materials are piling up. Every time I make a trip to Japan, I plan to write a lot during my plane trip across the Pacific Ocean and train trips between Tokyo and Osaka. But my resolution is usually completely destroyed by jet lag and my busy schedule. Then, I get more materials for writing as I meet more people and attend meetings and conferences. So I do not promise anything regarding writing blogs on this subject.

But let me say one thing. Unlimited bandwidth does not necessarily mean the waste of energy. I think ICT has been targeted as a culprit of waste because ICT is considered to spoil people by encouraging them to spend more electricity and energy than necessary. It is hard to assess the balance between the energy used by ICT and the benefits produced by ICT. But if the benefits far outweigh the energy consumption, ICT should not be blamed for the waste. It is a judgment call what a benefit is. For some, using Facebook and Twitter is a waste of time and energy. But for others like me, it is not.

For me, the idea of unlimited bandwidth brings up two thoughts. The first is that we will never accomplish unlimited bandwidth, because there is a finite limit to any resource we obtain. But here we should not take the term literally. It should mean we have adequate bandwidth to move us from the current status quo to the next level. The second thought is that this will increase data traffic, leading to even bigger Big Data that might accumulate for many years to come. Big Data is a good thing, but unnecessary and harmful Big Data may pose another environmental problem. I know no one seems to care about the second point now, but I predict that it will become an issue sometime soon. In any event, I will write more on this, if my unlimited bandwidth holds.

The subject of integrating facilities and IT data to monitor and control energy use at a data center is attracting a lot of attention. A talk by Kevin Malik, CIO of IO Data Center, titled "DCIM: The Future of the Data Center,” at the recent Uptime Institute Symposium, attracted a large audience with many people standing. Kevin said that IO developed their DCIM tool and applied it to more than 30 data centers. It worked fine, in spite of the many different building management systems involved.

I was always wondering how they integrated data from both sides. After talking to a guy at the IO Data Center booth and Kevin Brown of Schneider Electric, I think I finally get it.

Like a typical IT person, I still look at a data center from the viewpoint of IT but not facilities. What do IT folks do at a data center? They want to make sure that the applications necessary to conduct business run smoothly. Therefore, they need to make sure servers are running and other ICT equipment is in good working condition. So IT guys want to watch and monitor IT gear behavior. For that, SNMP and other IP-based protocols are used. But IT folks do not care or do not know how to pay attention to power consumption and other parameters, like voltage and frequency fluctuation, temperature, and power interruption, for the ICT gear.

On the other hand, facilities folks pay attention to power consumption/availability and adequate cooling and equipment for a data center. They care less about ICT gear behavior. So most of the time, DCIM with data from IT means power and temperature information about ICT gear via Intelligent Platform Management Interface (IPMI). You do not need to run an IP protocol for this interface to get what facilities folks want.

IPMI is defined by Wikipedia as:

System administrators can use IPMI messaging to monitor platform status (e.g. system temperatures, voltages, fans, power supplies and chassis intrusion); to query inventory information; to review hardware logs of out-of-range conditions; or to perform recovery procedures such as issuing requests from a remote console through the same connections e.g. system power-down and rebooting, or configuring watchdog timers. The standard also defines an alerting mechanism for the system to send a simple network management protocol (SNMP) platform event trap (PET).

So when most people talk about the integration of facilities and IT data at their data center, they are not talking about the integration of data relating to higher level stuff—like general server health, applications running on it, software utilization—that is usually a big deal for IT folks. In the SVLG DCEE conference held at the IBM Almaden Research Center last year, a session on DCIM was geared more towards IT. Cisco and IBM presented their systems to consolidate data from both IT and facilities. I am sure HP's OpenView integrates these two types of data.

But some vendors I talked to said that two types of data can be stored in a logical database that consists of two separate physical databases, one for facilities and the other for IT data. At the moment, this seems to be a good compromise. Looks like more research and work are necessary in this area.

Once, he arrived five minutes before his speaking engagement and delivered a good talk. He had just come home an hour before from India and sped from the airport to the conference site. I often go back and forth between the US and Japan, but I need at least a week to recover from jet lag. On the day I get back from Japan, I am completely beat and cannot do any meaningful work.

Dean delivers his talks well. Over the years, his speeches have become even better. He talks from his experience and gives really concrete examples. He must be one of the best speakers in the data center area, with charisma and a strong aura of persuasion. And I must add he is entertaining. Cloud computing, mobile computing, sensor networking, e-commerce, and others combined demand high density yet energy efficient data centers. There are many good speakers in each of these fields, but I think Dean is one of the few who can speak about all of them in context.

Deal Nelson

He started with the need for high density data centers at eBay. I talked about my experience with iPhone awhile back. 

If I did not have my iPhone, my reaction would have been different. People say that mobile computing, along with cloud computing and sensor networking, will increase Internet traffic substantially. Well, no one will disagree with that statement, but I cannot put my arms around that just yet, because I have a hard time finding real examples that are close to me or that I can relate to. I can draw an image of a vast number of mobile phones and other devices and a variety of sensors sending and receiving data constantly, but it is still theoretical.

Dean started with an example of a car sale on his eBay site. I must confess that I do not participate in any auction, whether online or offline. Maybe if I did, I would relate to his example even better. In any event, one of their users bought a high-priced car via eBay using an iPhone. I was sucked into the presentation, and so were other people in the audience. Some of his slides are available here. Also, a case study report is available from the Green Grid site.

He is used to how people buy expensive stuff on eBay, but this shocked even him. Yes, I am shocked as well. To start with, I do not have that kind of money for a car. Secondly, I am worried about the security of the eBay site (sorry, Dean, nothing personal; I am a very skeptical guy) and my iPhone security (sorry Apple, nothing personal, either). Either this guy has tons of money to lose or is a strong believer in e-commerce and its underlying technologies.

Well, maybe this guy is a very, very different kind of guy and an exception in doing e-commerce via mobile computing. In the next slide, Dean showed that this guy was an exception back in 2008 but not now or in the future. The next graph showed how people are changing their commerce behavior.

The figure partially hidden by a leaf for 2013 is 10T. One trillion is 1,000 billion.

"Web-influenced offline” means the actual purchase happens at a store, but before going to the store, a consumer researches online to decide which products to purchase. I think the online commerce trend will continue, but I do not think all the brick-and-mortar stores will be completely replaced by online shopping. There are some things people still want to see and touch before buying.

Having said that, when Amazon first started selling books, I did not think it would become this big. In the US, it is increasingly hard to find a book at a bookstore. Last year, Borders shut down. Some time ago, there was a small boutique computer store chain called Fat Brain (originally Computer Literacy Bookshops). I often went to the one at the Apple Computer headquarters. After the acquisition, it disappeared completely, in 2000. My choices got so limited. I stopped going to Barnes & Noble because I cannot find the book I want unless I order it. If I have to wait, Amazon seems to be a better choice.

 So much for my nostalgia for bookstores. To intercept this trend and exploit it, eBay did not sit still. It went on a buying spree for the necessary technologies, making 13 acquisitions in 12 months. It is a lot of work when you acquire and integrate other companies into your own. From human resources down to document formats and jargon, a lot needs to be changed. But when you consider 14 (13 plus eBay's original) computing systems, the effort to integrate them is a tremendous amount of work, let alone technology integration. Not sure exactly what eBay did, but some data centers might have been consolidated, and monitoring and other management probably used different tools and services. What a nightmare!

In a future blog, I will discuss how they built a new data center to go with these requirements.

Long, long ago, there was a debate on which gadgets to carry when you go out. The major contenders at the time were the cell phone and the PDA. In the IT field, many gadgets come and go. I wonder if the younger generations know what a PDA is. I may ask my younger colleague whether he knows what a PDA is. I remember at one of the Svforum meetings, a famous entrepreneur who came to the US from France with $100 in his pocket gave a talk on what to carry when you go out. He was quite successful and left his first venture then. He was Philippe Kahn and his first venture was Borland. His second company was named Starfish Software. In his talk, he emptied his pockets and took out about seven gadgets, including a cell phone and a PDA. I do not remember what the others were.

Time passed and many gadgets died away. I bought them and threw them away as they began to disappear from the market radar. I do not remember whether I buried them with honor, as in proper ewaste processing. There was no strong push or mention of ewaste processing at the time. It is only in the past several years that people have started talking about how to properly dispose of them. I wonder how much ewaste I personally produced. Proper burial with honor for my gadgets is not straightforward. I still have three or four old cell phones and several laptops and desktops with separate displays.

One thing is that I do not know how to erase all the personal and business information on the storage devices, such as the hard disks. If a computer is functioning, you can install a program to wipe out the disk contents completely. But if the computer is not alive (I mean if it’s physically dead), what can you do? My solution is to remove the hard disk and physically destroy it. That way, no one can restore the contents. I then put it back and take the computer to a nearby ewaste center, thanks to Apple computer. Then, why do I still have three desktops in my office? Well, I cannot get the hard disks out of the chassis. I can use brute force, but if I completely destroy it, will the ewaste processing place take it as a computer? So when I have time, I try taking it off. No success yet. I can take displays separately, but it is cumbersome to make extra trips.

Now am I to blame for my ignorance or my procrastinating personality? Do I need to be prosecuted by the Green Police? Since I cannot afford to hire an expensive attorney to defend me, I will defend myself. Here goes.

If I have violated anything that is not compatible with environmentally correct behavior, it is because:

1. Electronic gadgets manufacturers have not developed products that last longer.

2. Electronic gadgets manufacturers have not developed products that could be easily disposed of properly from the environmentally correct and security/privacy perspectives.

3. Society at large did not warn or educate me about environmentally correct behavior.

4. Some of my alleged violations are nullified by now because of statutory limitations.

5. If I am guilty, most of you are also guilty.

Am I convincing enough, Your Honor? Probably, #5 is a good argument because we need more prisons to accommodate the large population of violators.

Finally, I confess that I am computing-holic. I get depressed when my Internet connection is down, my PCs break, my iPhone does not reboot, and/or my email server is down. I need to be online at every minute, if not second, of my waking hours. Is this a disease or a crime?

It got much worse after I acquired an iPhone. Maybe I should blame Apple for my behavior.

In all the conferences related to data centers that I’ve attended so far, there is a session or two that discusses the integration of facilities and IT to manage a data center more effectively. So far, there does not seem to be an effective solution that might apply uniformly to the average data center. There may be a few reasons cited, such as no common management structure and the difference in culture and languages spoken.

On the former, Dean Nelson, VP of eBay, delivered a keynote speech at the recent Uptime Institute symposium in Santa Clara, California.

I will blog on his speech sometime in the future, but his presentation was very convincing because he started with the reason why more energy and energy density are required, with concrete examples and real numbers. In many discussions of ever-increasing energy and energy density requirements, no concrete examples are not provided. Skipping the details of Dean's talk, his answer for integrating facilities and IT was to put both departments under the same management. At eBay, they do just that. It is Dean who oversees both departments. This is not the first time Dean has done that. Back in the days of Sun, he played a similar role, keeping both departments under his control. This definitely helps. But as far as I know, not many organizations have done that (yet?).

For the second part, a common language plays a big role in culture. Coming from a two-byte language culture zone (Japan), it seems that European languages are very similar to one another. I am sure many people disagree with me. But the Japanese language does not even resemble Korean or Chinese. The grammar and characters are so different from any other language. If you give me several hours, I can tell you my sad stories about learning English. But let's forget about the difference talk. My point is even if you use English, your terminology matters. It is often said that even if you speak different natural languages, you can communicate well if you work in the same field and use the same jargon. Well, after doing some consulting work that bridges the US and Japan, I think this may or may not be true. But that is a different story I can talk about later.

In career counseling, experts suggest that if you want to change your area of focus, the first thing you should do is to pick up the jargon of the industry you plan to move to. Several years ago, I entered the green data center area and did just that. I read whatever was available, such as Rich Miller's datacenterknowledge.com, Dave O'Hara's , Green (Low Carbon) Data Center Blog and many APC white papers. It was fun to learn new things and use new jargon as if I knew something then.

During the Uptime Institute symposium, I sat down with a data center expert and chatted about this over a cup of coffee. He said that his background was in IT but he has picked up expertise in mechanical, electrical, and plumbing (MEP). He now speaks both MEP and IT languages fluently And this helps a lot to resolve any misunderstanding and conflict between facilities and IT when it comes to a data center operation.

Well, I speak standard Japanese, Osaka dialect, English, IT, and some facilities. I am getting pretty good at the facilities talk, but I need more practice. Probably the first thing to do is to make both facilities and IT people bilingual in both languages.

The conference was divided into very techie, hands-on workshops on actually hacking something on-the-fly and not so technical ones. It has been so many years since I hacked code, and I did not catch up with the conversations exchanged in the workshops, so I mostly attended keynote speeches and not so technical sessions. Other cloud conferences offer many different subjects, including markets and businesses, to attract a large audience. Yes, those things were discussed during the not so technical sessions, but there was something that was different from other nontechnical sessions in other conferences. I cannot say it very well, but my analysis is that the not so technical sessions were more technical than the general discussions of cloud computing in other conferences. Even though I thought I already knew the basics of cloud computing, many of the talks were geared towards technical folks, and I enjoyed them very much. In this blog and others to follow, I plan to summarize these discussions and inject my thoughts. I actually have written a blog on open source cloud computing. 

In this blog, I would like to concentrate on Gordon Haff's talk on cloud computing.

Gordon Haff of Red Hat

The stuff he covered was pretty basic, but I guarantee that you will appreciate what he said. It will give you an even better understanding of cloud computing. Here goes.

Gordon made five points about cloud computing.

Point 1

What is the fundamental difference between cloud computing and traditional enterprise IT? The next slide says it all, and I guess no explanation is necessary. Many people have made similar points, but this presentation summarizes the key points well. Remember that this is about public cloud. Public cloud went far in cultivating the cloud market, and private cloud is an attempt to duplicate that success in a private environment for enterprises. In a separate panel discussion, Nati Shalon of GigaSpaces summarized it very well. He said public cloud was all about productivity, while private cloud was about control. As the slide shows, public cloud enhances productivity in terms of agility and resource conservation (both cost and time). What is missing in this slide is control and security. Security is addressed in a following slide.

The control issue is sketched out in the following slide. Public cloud is great when we can ignore the issues listed on the left side, such as risk, compliance, guarantee, and differentiation points. In other words, enterprise adoption requires those issues to be resolved or at least for a compromise between risks and productivity to be reached after careful consideration.

Point 2

Cloud is fundamentally a computing model for data created by mobility, which is increasingly adding new types of data, such as context-aware data like locations and environments. Of course, stationary sensors placed at strategic locations of buildings, data centers, and smart grid systems also add to those data. The data that are generated, collected, and accumulated form Big Data, which is becoming a place for treasure hunting. Cloud is becoming the computing of choice for Big Data because of its traits, listed next to the cloud computing arrow in the following slide.

Of course, open source plays a large role in this, but even with a bunch of open source solutions for cloud computing, we still have very limited standardization. See my old post.

Automation, scalability, and self-service are all necessary for coping with this huge amount of data, or even to make heads or tails of it.

Point 3

Openness is a great concept for ICT practitioners. In a given enterprise, there are usually several computing entities invested in separately by different departments. To date, most of them are operated in a silo, independent of each other, and valuable data and information are not shared across departmental barriers. If the barriers are removed, all the investments, whether internal or external, can be used for the benefit of the entire enterprise.

This can be accomplished by open cloud. In the following slide, Gordon showed what is required for open cloud. Although I agree that open source is a big factor, it alone will not accomplish this. Take Linux as an example. In addition to a few dominant distributions, there are quite a large number of other Linux distributions, as indicated here. 

Each added its own utilities and libraries to the relatively small Linux core. If this is any indication, open source cloud will diverge in many directions with the small, standard core.

But don't think I am against open source. On the contrary, I am all for it. Points made in the two following slides are self-explanatory.

The existence of the community and ecosystems is important. The entire area of cloud computing is very wide and deep. No single product or service vendor can address all the issues and needs of the market. Therefore, it is very important to cultivate and grow the community and ecosystem as in other market areas.

Point 4

Security remains the largest inhibitor for enterprises to adopt public cloud. In the next slide, Gordon compared enterprise computing and cloud computing. Humans can perceive danger if it is visible, but we tend to ignore invisible threats because we cannot see them. We usually worry about perimeter security because an intruder who cannot pass the iron and cement gate cannot do any harm to our computing infrastructure. That is true, but in the age of ubiquitous connection, simply guarding the perimeter is not enough. We need to pay attention to network and other computing resource security.

Many analysts and experts say that the security available on cloud and hosting providers’ premises is usually much tighter than that of the typical enterprise. The reason given is that those who provide services focus on secure delivery because that is their core value proposition. On the other hand, enterprises worry about security, but securing their computing infrastructure is not their core value of business.

Gordon said that securing everything at an enterprise might give it some security and risk mitigation, but the resulting slow response to ever-changing circumstances might cause it to lose business opportunities.

Point 5

Finally, the following slide is great. Many people have discussed the progression of computing at enterprises. The following is a good way to understand it. Many people are still scratching their heads over the difference between a virtualized data center and a cloud data center.

Many people appreciate that virtualization revolutionized computing at data centers. It allows us to run multiple computing units (virtual machines, or VM) on a single server, cutting the number of necessary servers and increasing server utilization. At this stage, each server is independent and managed by a separate policy. This is a virtualized data center, but it is not a cloud data center. The next level up is a data center with holistic control and management of those virtualized servers with the same automation and policy.

Gordon further refined this model by adding applications and data management. This is where we are headed for. The current issues and interests are in applications (platform-as-a-service, or PaaS) and data.

Cloud people are busy with enhancing cloud functionality and other features like agility. I wonder whether cloud efficiency in terms of less resource consumption will become an issue anytime soon.

Mark Hinkle, senior director of the Citrix cloud computing community, gave a very good presentation on open source's role in cloud computing. As I’ve mentioned before, I was involved in open source by assisting with MySQL and JBoss, and I’ve always wondered whether open source and cloud ever got closer together. I knew of some movement in this direction, like Xen, but recently open source has played a much bigger role in cloud computing. And Mark's talk was very timely. I will touch upon software efficiency, and thus its energy efficiency, in a future blog. It’s interesting to think about software efficiency from the standpoint of open source.

In any event, Mark's talk was interesting because he surveyed the solutions that are available for cloud computing at this time. As you can appreciate, some open source solutions come and others go. One of Mark’s slides indicated that Red Hat's OpenShift was almost released as open source. But he acknowledged that Red Hat officially made it open source only a few days ago, after he made the slides. It is very useful to know what is currently available. It is not an easy task to keep the list up to date. For that reason, Mark's work is very useful. I summarized his talk below by using some of his slides. His entire presentation is available here. http://www.slideshare.net/socializedsoftware

For those who are not very familiar with open source, its merits are these:

I mentioned that an open system can be established as long as a standard exists. Also, even if an open source solution exists, like Xen, the current clouds are not open, as Mark pointed out here:

Nonstandard virtual machine (VM) file formats prevent VMs from being moved around from one cloud to another. Amazon, which dominates the public cloud, uses its own file format, which is not compatible with the original Xen file format. Having multiple file formats is not very efficient from many points of view. One organization may need to maintain multiple file formats and may waste human, equipment, and time resources. That might be translated into cost and energy use.

In the next few slides, Mark surveyed the solutions that exist for each category of cloud computing, starting with tools to create VM appliances. There are many more solutions, but only those with high traction are listed. Also, each package keeps changing, and you can click on each in his original slides in http://www.slideshare.net/socializedsoftware for the latest information.

Then hypervisors are listed. Hypervisors were the beginning of cloud computing and provided virtualization. As Gordon Haff, another speaker, mentioned, virtualization is an important piece for cloud computing, but by itself does not constitute cloud.

The next slide lists infrastructure-as-a-service (IaaS) solutions:

Of these, I am not sure how much traction OpenNebula is getting. CloudStack was recently released from Citrix as open source. Citrix is also supporting OpenStack, and I should have asked Mark how Citrix plays both cards at the same time.

Next comes storage:

Platform-as-a-service (PaaS) is gaining popularity:

Infrastructure is important, but we need tools to manage such infrastructures. These are the categories:

Tools for each category will be presented next. Open source provisioning tools are here:

Configuration tools:

Monitoring tools:

I once talked to GroundWork, which is based on Nagios. http://www.gwos.com/ They provide an open source and a commercial version of their monitoring tools.

Automation and orchestration tools are here:

The next, very helpful slide shows how these tools interact.

Finally, I want to thank Mark for his useful presentation. The open source cloud scene changes again.

Before getting to the interview, let's see how Hadoop is doing. The Hadoop site is a good source of information. It contains information related to the Hadoop project as well. We all know that Hadoop is going strong and getting stronger every day as the savior of Big Data solutions. Alex Handy of SD Times wrote a timely article on the Hadoop platform (URL),　Hadoop 1.0 IS All About Hadoop 2.0” You can read the article, but these are its two major points:

• The version known as 0.22 was relabeled as 1.0 on January 4, 2012.

• Version 0.23 is speculated to be 2.0.

Version 1.0 includes the following changes:

• Major upgrades for HBase

• Performance enhancements

• Bug fixes

HBase is the in-Hadoop SQL database to which the user can import data for analysis by Hadoop.

Version 0.23 is speculated to be 2.0. One of the major tasks left to do after version 1.0 was to make HDFS High Availability (HA), the underlying file system for HBase. Hadoop 2.0 is a complete rewrite of its core, known as MR2 (Hadoop map/reduce). After the rewrite, MR2 will become a general scheduler and resource manager. Another feature of Hadoop 2.0 is higher scalability. The article is a good summary of what's happening with Hadoop now and in the near future.

Let's shift gears. Some people wonder if the market can sustain several Hadoop companies, when big shots IBM, Oracle, and Microsoft have entered the market, along with startups like Cloudera, Hortonworks, and MapR,.. If Zettaset is yet another Hadoop company, can they survive? It is a very interesting question, so I arranged an interview with Brian. Initially, I was not clear about what they do and how their solutions relate to Hadoop and other business Hadoop companies like Cloudera, Hortonworks, and MapR. The video interview (given at the end of this blog) runs a little less than three minutes, and in it you can find his answer to those questions.

These are two of the takeaways from the video:

1. Zettaset enhances Hadoop environments by wrapping raw Hadoop. Through this wrapper multiple pieces, ZooKeeper, MapReduce, and HDFS, are well orchestrated with Hadoop.

2. In conjunction with the first point, security is also provided.

The first point is well depicted in the figure below.

Extra functions provided by wrapping Hadoop by Zettaset. (Source: Zettaset)

This is done by tapping into the APIs provided by Hadoop without altering the Hadoop source code. As you can see, and as Brian emphasized, they do not want to fork the code. Brian explained that by using the original code without modifications, their product can wrap any Hadoop system, business enhanced or not. So Zettaset can provide solutions to any of the Hadoop products. That means they can wrap the original Hadoop and versions by such providers as Cloudera, Hortonworks, and MapR. This is a very wise way of providing solutions without betting on any particular business solution provider.

I did not ask Brian whether the following case works. Let’s say that one of the business-enhanced versions uses its own API (and let's call it enhanced API, for the sake of reference) based on the original API. If Zettaset does not have a hook to this enhanced API, it cannot provide its enhanced functions. This needs further investigation.

Here's the video interview.

RSA 2012 – Zettaset’s Brian Christian On Management Platform And Secure Wrap around Layer For Hadoop And Big Data

Finally, I have to say something about energy efficiency to make this blog relevant to my theme, the intersection between ICT and energy. A vast amount of Big Data is unmanageable without a solution like Hadoop. Zettaset makes it more manageable by providing good management and security by wrapping any Hadoop system, business enhanced or not. I have a wild idea. If Hadoop and other Big Data solutions can identify useful data and derive good information from it, why not identify what is not useful in the huge amount of data and flag it for later deletion, or at least combine, condense, and reduce it into a smaller version? If we know what is not useful to retain for the future, then there is no reason to keep those data, and we can reclaim their storage areas for other use. It is not trivial to classify some data as not useful, because depending upon the point of view, the same data set may or may not be useful. Some cases may be obvious. By removing unnecessary data from storage, we could increase energy efficiency.

At the recent RSA conference, as I was rushing to the next appointment for a video interview with Akamai on the show floor, someone called my name. The crew I was with were walking ahead of me and the voice came from nearby. It was Larry Stefonic. Larry was senior VP sales for MySQL and, later, president of MySQL Japan, a vastly successful open source SQL database company. As you know, it was acquired by Sun, and Sun was acquired by Oracle. Therefore, MySQL is now part of Oracle. Larry hired me as a market scout for Japan. I tapped into the Japanese market, including promotion, business development, and sales lead generation for him.

While doing that, I learned a lot about open source and its merits. I also did similar work for JBoss at about the same time. Then, both MySQL and JBoss got sold and Larry disappeared. With that, I lost interest in open source for the moment and moved on. After I reconnected with Larry via LinkedIn, I realized he was in Montana and had started a new venture called yaSSL (yet another SSL), which is based there.  I knew SSL had become the standard for web browsers and small-footprint embedded systems. From 1995 to 1998, I was running a security product (firewall and VPN) business at an international company, so I have a good understanding of security concerns, such as filtering, authentication, authorization and encryption. But both security and open source were behind me, and I did not bother to take a close look at his company. Occasionally, I would see his posts in LinkedIn and Facebook and knew his company was doing fine.

Sorry about the long introduction before the meat of the discussion of yaSSL. It is an open source security (SSL) product company. A short but good summary of yaSSL is given here. For the SSL tutorial, go to the yaSSL site.  yaSSL is a combination open source and security company. Understanding it is a piece of cake for me. Its business model is very similar to what MySQL had for sure and has now. They provide their SSL products with two licenses: GPLv2 and commercial. If you do not know what that means, you can find an explanation on their site.  But in a nutshell, the commercial version is the traditional commercial license deal, while GPLv2 allows you to make a copy of the source (product) and any number of changes to the product, as long as you contribute your changes back to the project.

The advantages of open source over closed source are often cited. I used to listen to Larry's pitches on their merits. A partial list includes:

• Inexpensive because no license fee

• Easy to adapt to your own environment

• Market needs quickly reflected by community response

• Quick support by the community

• Debugged by a large number of community eyeballs

See four gentlemen whom I met at their booth at the RSA conference. yaSSL was founded by two people, Larry Stefonic and Todd Ouska, both pictured below with early employees John Safranke and Chris Conlon.

From left: John Safranek, Todd Ouska (cofounder), Larry Stefonic (cofounder), and Chris Conlon .

Their products are also geared towards embedded systems. Yes, I did embedded systems as well and know how tight their requirements are for executable footprint and memory sizes. Their products are listed here:

• CyaSSL is a C-language-based lightweight SSL library targeted for embedded and RTOS environments, primarily because of its small size, speed, portability, and feature set.

• yaSSL is a C++-language-based SSL library for developers more comfortable with that language.

• yaSSL Embedded Web Server is a fast, embeddable, and easy-to-configure web server with a strong focus on portability and security. It offers HTTPS support built in through the CyaSSL embedded SSL Library.

• yaSSH is a small, portable SSH implementation currently in its early phases targeted for use by embedded systems developers.

SSL and its improved version, Transport Layer Security, are IETF standards now.  There are, naturally, differences among the different versions of SSL and TLS. yaSSL supports SSL 3 as well as TLS 1.0, 1.1, and 1.2. Many copies exist over the Internet, and they are compatible with all the major versions.

Their competitive messages are well described on their booth banners.

There is a Wiki page for comparison of available TLS solutions, open source or not, in which yaSSL is listed along with open source and closed source solutions. 

Among the open source versions, CyaSSL (C version) is often compared with OpenSSL. You can find the comparison by contacting them at info@yassl.com. However, one thing that caught my attention is their footprint size. It ranges between 30 and 100 kB, and the runtime memory use ranges from 3 to 36 kB. This is very small. With these resource requirements, their product can be integrated into a tiny wireless sensor, and such sensors are considered the new culprits causing Big Data. As in any data transmission, data by those sensors must be encrypted for security reasons, especially when used for strategically important infrastructures like the power grid. With good security available, such sensors will be deployed more widely.

It was imperative that they make their software smaller to fit into a tiny sensor with limited power. I assume they used every method available for size reduction. When we look at current software development for larger systems like servers and workstations, including PCs and laptops, we take it for granted that infinite hardware resources are available, and convenience rather than efficiency or performance dictates our choices. Most web languages are script rather than compiled, taking more resources to execute. Executing script languages generally requires more powerful processors and more memory, and those consume more resources.

Green IT means two things: greening IT and greening other things by IT. Most discussions center on making hardware greener, but not software. If we can design and implement smaller and more efficient software with an algorithm that minimizes resource use, we can keep energy consumption to a minimum. I am not advocating that we write software in machine or assembly languages. But we can do a better job of developing software for energy efficiency. Because we throw more resources at convenience, I feel programming in nonembedded systems has lost a sense of efficiency. We can only attain so much energy efficiency by working simply on hardware alone. It is about time to pay more attention to software for energy efficiency. I am not talking about making other things more energy efficient but software itself. More discussion of this to come.

Crowd at the registration area.

It goes without saying that security plays a great role in almost every product segment. This industry is growing even though many technologies come and go. Two noteworthy trends are mobile and ID management. Mobile, along with cloud computing, is one of the few areas where IT shines. Security of mobile computing means different things to different people. A company like TeleSign uses a mobile phone to verify identity for network access, while Mobile Active Defense coined the term mobile device security management (MDSM), claiming that the security services covered by the blanket term mobile device management (MDM) do not protect your mobile phones enough. In addition, Echoworx provides an encryption technology for mobile phones like the iPhone. I was behind the curve in not knowing how powerful the iPhone’s processor is. It has enough power to encrypt/decrypt on its own platform. And their app can be downloaded OTA easily. They allow encryption on both communications and data storage. It is interesting to observe that people who had been using a cell phone and moved to a smartphone do not seem to realize a smartphone is really a small computer with all kinds of security breach possibilities. Those of us who came from server and PC environments may feel constrained on such a small platform, but we are well aware of the security threats of mobile computing.

The second noteworthy trend I saw, ID management, is of vital interest. We get online using multiple IDs to access many different sites. I am sure I have more than 20 IDs of my own for work and personal use. I, for one, cannot remember all the IDs and their associated passwords. It would be a nightmare for an enterprise with many employees having multiple online accounts for different clouds to manage such a huge number of IDs. As we talk about hybrid clouds, people want to have a single ID to seamlessly access any cloud they use. BeyondTrust provides privileged ID technology applicable to physical and virtualized environments, including cloud computing. I did not have a chance to drop by Symplified, but they seem to have a good solution.

We need a robot at any show!

Other companies, like Fortinet, Solera Networks, MokaFive, Akamai, and Zettaset, were also very interesting. I plan to blog on some of the companies mentioned here. Fortinet provides unified threat management (UTM) appliances. In the mid-1990s, I ran a division of an international company that developed and marketed a turnkey firewall and VPN box (a UTM appliance, in today's terms), running a PC version of Unix (no, it was not Linux; there was such a thing then). Now hacking is more sophisticated and network speed is reaching 40 G and 100 G. Affordable yet easy-to-use appliance solutions are a must for any size business. Solera Networks has a technology to monitor high-speed network flows and detect anomalies by using their analytics. Andrew Brandt was very knowledgeable. We retook the interview because the first was a bit too long and too detailed. Actually, I liked the first better and appreciate his profound knowledge. MokaFive was founded by four researchers from Stanford University and Vinod Khosla, who makes five. Virtualization on the desktop is behind server virtualization. Incidentally, the server virtualization pioneer, VMware, has its roots at Stanford. Mendel Rosenblum is an associate professor there and one of the founders of VMware. They provide a copy of the virtualized environment on a central server, and each desktop runs it. In this way, they avoid complex agents and copying each update from the server. Can they be the next Sun, Yahoo, VMware, or Google? That remains to be seen.

Unfortunately, I did not hear a word about energy conservation or technologies along that line. Only remotely related was yaSSL, which provides security (SSL) for embedded environments. The embedded platform by necessity makes their footprint and resource requirements extremely small. We are talking K not M or G at all. If we are consciously saving the consumption of resources in embedded environments, why not apply the same mental frame to other platforms? I would like to revisit the idea of Energy Star for software, which no one gets except Jose Iglesias of Symantec.

Finally, my sincere apology to the GreenSQL people. Because of our internal miscommunication, we missed an appointment with them. They are a SQL firewall company, and their product protects MySQL and Postgres databases specifically. As hacking and attacks get more sophisticated, we may need a specialized firewall for each application. I do not know, but something like a SQL injection attack may be hard for firewalls to detect at corporate entry points.

More to come. Stay tuned!

This was my second conference since getting my iPhone. It was very handy to have a smartphone on the show floor, but that isn’t what I plan to talk about here.

Instead, I want to share my observations on how people conducted their business as they walked around the huge Moscone Center.

Enterprise vs. startup environments: Like any modern conference, this one provided Wi-Fi access to press and attendees. At the recent Cloud Connect, access was open, without any encryption or pass key. Because this one was an RSA conference, Wi-Fi access was controlled with WPA-2 Enterprise. My Windows environment is somewhat special in the US because it is a Japanese version of Windows 7. The RSA conference people prepared instructions for accessing the Wi-Fi network. The configuration of Windows in the Japanese version is not the same as in the English counterpart. After a few attempts, I gave up, went to the press room, and connected to one of the good old Ethernet cables there. The conference people had set up a wireless helpdesk to give a hand to poor souls like me. But I did not go there, to avoid the hassle of working with the helpdesk staff. So throughout the conference I was online by wire.

This is not meant to be a criticism of the conference. The RSA conference represents established enterprises, while Cloud Connect represented the startup culture. Security is a big concern for enterprises. I thought the difference between the two conferences was a good contrast.

Mobility, Wi-Fi, ubiquitous, and battery: My iPhone loses power rapidly and requires frequent recharging, so I bought another gadget that is very thin and mounts on the iPhone for additional battery capacity. During the conference I saw people charging their iPhones constantly everywhere, and PCs and Macs were no exceptions—people were charging those laptops all the time and everywhere. This is definitely killing ubiquitous access to the Net. Mobility or wireless is powerless when you cannot run your gadget because of power loss.

Also, I wonder how much power was consumed by smartphones and laptops alone during the conference. I do not know the precise number of attendees, but my guess is 2,000 to 3,000. And almost everyone had either a smartphone or a laptop or both. Also, even if the battery technology is improving, losing some power during charging and discharging is inevitable. At this scale, each loss and the total number of devices are still manageable, and the loss does not amount to anything big. But if we are talking about millions and billions of these gadgets worldwide, the total loss can be staggering. What with Big Data, I wonder if the next energy crisis will be caused by ICT.

Internet growth is and will be sustained by mobile technologies and wireless sensors. Wireless sensors are stationary, while mobile phones are truly mobile. Laptop computers with Wi-Fi and cellular connections are mobile but a little different from mobile phones. It is likely that you sit down someplace to work on your laptop computer, even if it is wirelessly connected, but you may conduct your business with a mobile phone while moving.

Network access and battery technology hold even greater growth potential. The iPhone design is great for maximizing a connection to a Wi-Fi access point, which often is free and faster than a cellular connection, which charges on the basis of data traffic volume. As far as I can tell, the iPhone does not have a very sophisticated configuration for encryption or authentication, which enterprises require for connection to a Wi-Fi network.

If these two problems of network connectivity and battery life are solved, market growth will skyrocket. I have no idea where and how far this is going. We have been working on ICT to make our lives more convenient, commoditizing much of the equipment and making it cheap to use. It is a good thing for convenience, and convenience fuels further growth in ICT and its application markets.

What is orthogonal to this trend is security, and people are aware of it. Free information flow in the Internet is a good thing, but bad guys are also free to come in and poison the water. Because of this, we need firewalls and malware prevention. We need sophisticated Wi-Fi access, as at the RSA conference. On one hand, security dampens growth, but on the other hand, new technologies and business opportunities are born to implement security, as we saw at the RSA conference.

What people are not paying attention to is energy use. The online growth trend will continue, and it is probably impossible to stop it. Because it is cheap to use, vendors throw in any amount of email storage like Gmail and Hotmail for free. The providers are happy to accommodate the demands because it is a good promotion for their companies. People waste storage and other IT resources because they seem to be infinitely available forever. When gasoline was only 50 cents per gallon in the 1970s, people drove large automobiles and did not pay attention to conserving energy at all. Forty years later, we face $5 gasoline and worry about depleting the oil supply. Maybe a power shortage caused by ICT fueled by mobility won’t happen for another 40 to 50 years. I certainly hope my fear is groundless.

Different people use different methods to sort out the information. I cheat a little bit by pinging experts for their opinions. At Cloud Connect 2012, I had an opportunity to catch both Mark Thiele and Tim Crawford for interviews. Even though the interviews were short, I got a very good summary of the state of cloud computing.

I asked Mark to list the three biggest trends. Here's his top three:

3. Management of cloud resources with some risks of building large-scale infrastructures rapidly, involving things like networking and licensing.

2. Organizational changes to accommodate new environments caused by adopting cloud computing.

1. Acceleration of cloud computing growth by the mobile segment and opportunities for business.

See the actual interview here (roughly three minutes):

Questions to Tim were regarding new IT and the CIO’s role when cloud computing is adopted. A short summary follows.

• Cloud computing and the consumerization of IT are a blessing to IT and not a curse.

• Changes that involve technology processes and organizations are hard to accept.

• IT can take on a new role as a knowledge worker by outsourcing noncore functions to outside providers.

• Although the changes will take some time to happen, after they have, IT will finally be a strategic enterprise weapon.

See the actual interview here (roughly three minutes):

Both Mark and Tim said that cloud computing will change IT substantially. The changes are not just in the technologies but also in how work is done (process) and how departments are structured. After all, cloud computing has a lot to do with operations. Both experts said that after the changes, IT will be reborn into a new stage that will be a welcome change.

Mark’s points about cloud management and mobile are very noteworthy. As cloud evolves from public to private and then to hybrid, many more variations and combinations of clouds are formed. Some private clouds may be on-premises but others may be off-premises. People may want to mix and match several public and private clouds at the same time. This will require good cloud and resources management.

The mobile market is rapidly growing, which accelerates cloud computing.

I have interviewed more people from several companies and will write about them in future blogs.

In any event, after getting an iPhone, I now have firsthand experience with mobile computing and mobile clouds. Now I know why this is such a big deal. Those who were already using a smartphone would tell me that this is yesterday's news. Anyway, I have a lot of catching up to do. But not everything about mobile computing is great and dandy. Here are some of my complaints:

1. Short battery life: It barely lasts one full day and I need to keep recharging it.

2. Soft keyboard: The soft keyboard is not the easiest thing in the world to use. The voice-activated agent, Siri, is good but does not always work well.

3. Slow connection: Without help from a local Wi-Fi connection, my 3G connection is a little too slow to browse websites. Verizon has an optional data plan to use the iPhone as a hotspot. Initially, I thought it was a good idea, but the slowness of the 3G connection’s speed in getting me onto the Internet may test my patience.

4. Energy waste: The iPhone ecosystem (equally true of the Android’s) is designed to encourage the user to enjoy services, which is translated into energy and cash use. People may use them mainly because they are fun, and without regard for energy consumption. Some of the services may not be necessary, leading to the abuse of energy.

Still, those problems, except the last, will eventually be solved, and I see mobile computing changing the world and ICT as we know them. However, I do not have a solution for the last problem. People do not seem to get it when I talk about the potential for energy abuse by mobile computing. You’d think I just stepped out of an alien spaceship.

In any event, I will discuss what I heard about mobile and cloud computing at the recent Cloud Connect 2012 in the next few blogs. There were many interesting subjects covered in the conference, like those below.

• Private clouds

• Legacy applications in the era of cloud computing

• Mobile

• Big Data

• Cloud management

• Open source

There was a track on mobile computing and mobile clouds on the last day, which was quite interesting. By the way, all the keynote speeches are available at tv.cloudconnectevent.com, regardless of your attendance. Unfortunately, presentation materials are available for attendees only. Stay tuned.

It was simple in the enterprise world. There was a main business organization with a handful of supporting departments like HR, accounting, legal, and facilities management. There was no doubt that the business organization was king. It ruled the enterprise without question. Then computers came to the enterprise. At the beginning, computers were subordinate, simply supporting business. As computer technologies advanced, it became necessary to form an independent organization called IT. Even so, IT was subordinate to business. Business dictated and IT followed.

It is hard to say when, but at some point IT got so powerful that business could not tell it what to do. Or more accurately, business was still boss, but IT became less responsive to business needs, and not just to requests; IT did not and could not respond to the changes taking place around it. IT was supposed to support and accelerate business goals, but it became a barrier to business. It became so bad that people said IT was the place where big, important projects went to die, according to Tim. For some time, business was frustrated with IT but did not have the means to bypass it. Then came the era of cloud computing and commercialization of IT. Business secretly formed a shadow IT department, like a shadow cabinet in the UK, and started to bypass IT whenever possible. Who can blame non-IT folks who need IT services yesterday? If it is going to be months before IT can satisfy my needs when I need it now, I will bring in my own gear or outsource the services. The big difference now is that we can do it if we want. It was not possible to do so only a few years ago. Tim said some CIOs now realize this and are working to face this straight, but many CIOs still think the old and traditional way of running IT departments is appropriate.

There is no real department for shadow IT. In a way, any business or non-IT staff who needs IT services can virtually form a shadow IT group and outsource their needs to cloud and mobile computing. If there were really one physical shadow IT department, it would be easier for the real or traditional IT department to confront it and take back control. But this shadow IT group is like guerrilla warfare. There is no particular place the group shows up. It appears where there is a real need for IT services, gets them quickly, and then may disappear. As far as I know, Tim is the first person in IT to admit that the blame is on IT for bringing this situation on itself.

This may not be a good analogy, but at the same conference Pascal Finette of Mozilla gave a keynote speech on open innovation. His theme was that opening the barrier could accomplish even greater results. Maybe it is a stretch, but I am saying that the IT department should be more OPEN to what their internal customers want and work with them for the entire company. Tim said that CIOs in this new era should take a very hard look at their core capabilities, review their portfolio of services, and decide what should be retained and what should be outsourced. This is a hard thing to do because it may mean downsizing the IT department. The CIO and the IT group need to have heart-to-heart discussions among themselves and with their internal customers.

It is easy to propose this at a high level, but how do you actually implement it? When it comes to cloud computing, the number of popular services like Salesforce.com and AWS is low, and it may be easy to evaluate the usefulness of each. But if IT approves a "bring your own device” (BYOD) policy, the sheer number of variations could be a problem. I would like to ask Tim about this. Maybe he has already addressed this in his blog.

Tim advocates that IT take the initiative to evaluate current services and gear before its internal customers come asking for support and approval. It would be great if, when a customer asks for services or new technologies, IT is ready to embrace the request and seamlessly integrate it into the current portfolio of services. That is a new value to IT and will definitely increase their importance in the company. There is going to be new pressure on IT. IT needs to study the market trends and new services and technologies daily and constantly. This alone would produce a ton of work. So this makes it even more important to review what they currently have and remove services of low priority from their portfolios. If some services can be easily outsourced without losing important elements, such as security, they should be.

For example, if IT studied Salesforce.com well in advance of a customer request to incorporate its services for business, it could accommodate the request with strong support. Moreover, because IT is probably the only department that interacts with most, if not all, departments in the company, it can facilitate communication between departments. For example, if IT understands how marketing and sales operate, it could avoid sending conflicting messages to the same customer from marketing and sales, which is very embarrassing, according to Tim.

What if IT provides virtualization? Would it be enough to prevent internal customers from resorting to outside clouds? Virtualization and cloud are two different things. I caught Tim after the interview to get more details on this, but that is a story for later. For now, let's say virtualization is not cloud, and cloud is more than that. Tim said that the three pillars of cloud computing are economic value, flexibility, and responsiveness. In most discussions, economy is considered the key value for cloud computing, but Tim said it might not be the most important factor. Sometimes cloud computing may cost you more, but you may want to adopt it for its flexibility and responsiveness. To sell cloud computing to your CFO and CEO, you should be able to make each point by giving a concrete example and the savings associated with it. For example, scaling from 50 to 1,500 servers over days and weeks could only be accomplished via cloud computing; no traditional methods accommodate change on such a scale. Moreover, you should be able to convert this to $ to explain it to the CFO and CEO.

Is virtualization a necessary process before adopting a cloud service? Tim said that the adoption of virtualization is probably in the 30% to 40% range, less than the 50%-plus estimated by IDC. If you have not virtualized your data center yet, you need to take a holistic view rather than simply considering virtualization. Looking solely at technology (virtualization) is not right. After all, virtualization maximizes server utilization, but other things—like maximization of the organization’s resources and processes—are also very important. So, according to Tim, it is OK to bypass virtualization and adopt cloud services by paying attention to the whole picture.

What about organizations that have already invested in hardware? It is easy to talk about startups, which have very little IT gear, but larger companies, even small to midsize companies, may have a hard time moving to a new paradigm. Tim's answer was clear: whether it’s hardware or software, everything will be replaced at some point. The lifespan of server hardware is about four to five years, and software applications must be upgraded or replaced at some point. That is the time to take a hard look at what to do from several points of view: economy, technology, process change, and the whole organization.

Finally, security is the number one inhibitor mentioned in many research results. The perception is that only larger companies need to consider both security and regulatory conformance and that small startups can adopt cloud computing readily. Tim said both considerations are required in any business of any size. So that kind of argument does not have merit. Also, he said, in many instances cloud providers' data centers and security measures for other elements are better than those of companies of many sizes and kinds. In any event, if you worry about data security, you need to weigh convenience and ease of use with security risks. Having data at your own data center does not make it more secure than the data hosted somewhere in clouds.

There is a lot of talk about cloud computing, but sometimes people duck the hard questions. I will catch Tim in the future to discuss some of mine:

• How do you transition from virtualization to cloud computing?

• What are public, private, hybrid, and federated clouds?

• Is there such a thing as private cloud?

• Hybrid clouds? What about interoperability of VMs between private and public clouds?

• What new requirements, such as ID management, do federated clouds bring?

• What are green clouds? Can you quantify the greenness of clouds?

So stay tuned!

At Teladata's recent Technology Convergence Conference, Andrew Dailey of MGI Research interviewed him.

From left: Andrew Dailey and Tim Crawford have a fireside chat.

I will write a few blogs based on this session. The questions and answers below are edited, and my notes are indicated by my initials in parentheses (ZK).

Only the fittest will survive in the technology and business world, as in nature. Andrew’s first two questions were pretty blunt and provocative. The first was whether we still need a CIO. Tim's answer was yes and no.

(ZK) People in a corporation can bypass IT and get necessary computing bandwidth with their credit cards from cloud vendors almost instantly, without a lengthy request, review, and approval process. They do not need to purchase new IT gear to satisfy their needs, and they charge the cost as part of business costs rather than capital expenditure.

Also, people can bring in their own gear, such as smart phones and tablet computers. Some may even bring in a Wi-Fi-access device for their convenience. Smart phones, such as iPhones and Androids, and tablet computers, such as the iPad, hit the consumer market first. Even before corporate IT departments embraced and supported such consumer-oriented devices, workers brought them in and used them, bypassing IT departments. With this gear, people even form personal clouds on top of private and public clouds. If people can do their jobs easier and quicker without going through their IT department, we do not need the IT department or consequently the CIO.

Tim thinks in spite of this trend, there should be someone who decides and shows the direction for corporate IT.

(ZK) Each individual can do whatever they want to do with their own gear and outside vendors. But if there is no clear vision of where their IT is headed, there is just going to be chaos.

Tim said that today's CIO needs to change with the current mindset to accommodate the sea change in the corporate environment.

(ZK) This is very true. If the IT department remains a place where they can close their eyes and ears to change, neither it nor its CIO is necessary.

Andrew’s second question was whether each company needs to build, maintain, and run its own data center. Tim’s answer was very clear. If the data center is not your core value for staying competitive in your own field, you should not get involved in data center operations.

(ZK) When a provider of data center services, such as colocation or wholesale, claims that you should outsource your data center services, they cite several reasons: construction, maintenance, and operation require a large capital outlay; you must acquire and retain expert talent; and there will be a time lag until actual use. Even though this is true, because it is coming from a vendor, you may not buy it with 100% confidence. But coming from an independent source, it is more convincing. Still, it is not easy to determine whether you want to move all your gear out of your data center to an outside provider. Also, what about private vs. public? Hybrid clouds? I will catch Tim later and ask him more about it.

There were many other interesting conversations during the session that I will share with you in the future. Meanwhile, you can catch more of Tim’s thoughts on his blog.

Pascal Finette of Mozilla

I never read Mozilla's mission statement before, so I visited their home page to do so. It says:

Mozilla is a proudly non-profit organization dedicated to keeping the power of the Web in people's hands. We're a global community of users, contributors, and developers, working to innovate on your behalf. When you use Firefox, you become a part of that community, helping us build a brighter future for the Web.

A good video (a little over three minutes) describing it is here.

Throughout his talk, Pascal used a Star Wars metaphor of the fight between the evil Empire, which wants to monopolize the market and maintain the status quo, and the Rebels, who want to bring innovation to the status quo. He started by saying that openness is happening everywhere, including in the taking down of the wall in Berlin and in the Arab Spring. Also, new industries emerged in snowboarding, mountain bikes, and hospital instruments because ordinary people started the initial idea and other ordinary people signed on to join them. Other examples include a cloud backup company, Facebook's Open Compute project, and Openstack.

Pascal also said many smart people work for other people. So how can you pick their brains and get their collaboration? How can we combine their brain power and knowledge without hiring them? Knowledge is sticky. What that means is that knowledge is hard to distribute. We need to combine the knowledge scattered among many smart people to form a big brain to solve new problems. To create innovation, we need to create a lot of ideas first and funnel them down to a small number of ideas that could actually make it. The biggest innovation problem is the many barriers that prevent the initial idea from forming. In the following picture, Pascal showed four quadrants to indicate where community and collaboration from the mass is situated.

Then, he used a web browser as an example to show how open innovation has progressed. He described the web browser war as fought in three time periods (Act 1 to Act 3; remember his Star Wars metaphor). Act 1 includes the birth of the web browser and the emergence of a dominant company, namely Microsoft. In 1994 Netscape released the first Netscape Navigator. In 1998, a full-fledged browser war began. Microsoft gave out its browser for free by bundling it with the Windows OS. Netscape countered by making its browser code open-sourced. Later that year, AOL bought Netscape. By 2001 Microsoft's Internet Explorer 6.0 had a market share of 98%. Microsoft felt that the browser war was won and allocated only two people to support it. In Act 2, in 2004 about 20 people got together and released Firefox 1.0. They did not have expertise in marketing or money. The community rallied behind them to provide money and even marketing ideas, including ad graphics. With community support, they could put a two-page ad in the New York Times.

The launch campaign was a big success. In version 2.0, they even used a farmland design to draw attention (see below).

Then, comes Act 3, today. The community is still very active in supporting Firefox, writing roughly 40% of its code. Its Italian community projected an image on top of the Coliseum in Rome to draw attention to the browser. (It drew frowns from the Italian police.) Its market share is about 25%.

Speaking of the international community, the original English version is now available in 84 languages, even in an obscure Spanish dialect that it is the only web browser to support. I use both English and Japanese versions of Firefox daily.

Back in 1994, I was with a company in Silicon Valley that was a subsidiary of a Japanese company, and I was told to engage in a license agreement with Spyglass, which had exclusive rights to the original code that was developed at the University of Illinois at Urban-Champaign. I was also involved in a negotiation with Netscape and met with its two founders. I just want to mention that I have been using Firefox since its inception. I use Internet Explorer when absolutely necessary, though.

Pascal also told us the five things you must do to achieve success in open innovation. I’ve added my notes (ZK).

Here’s what to do:

1. Make your products superior. t. (ZK: Only great ideas excite people to come work with you. Products that are not superior do not deserve other people's time, including your own.)

2. Push most decision-making to the edges. (ZK: VISA is a federation of many banks and other financial organizations. They are a combination of chaos and organization. Each Firefox module is assigned to a team leader, who is not necessarily hired at Mozilla. Only the most qualified people make decisions, regardless of their affiliation. This is scary.)

3. Communication is a key. (ZK: Internally communicated information is not visible to the community. Mozilla team meetings are open to anyone who wants to participate.)

4. Make it easy. (ZK: They make it easy to help them. If someone has only five minutes to help you, get his help. If it is two hours, accommodate their constraints and get the help.)

5. Communities are not markets, but members are citizens. (ZK: The community consists of citizens like you, and you need to acknowledge that each citizen has the same right as you and keep in mind to work with the community rather than exploit it.)

I can relate to these points well because I got involved with MySQL and Jboss, which are known as business open-source companies. Granted that not all companies can adopt this idea, the trend seems to be moving from closed intellectual property (IP)to shared open ideas. What's next? How do you differentiate? It appears to be in the services around the products and ideas. Hardware got very cheap because it became a commodity, while most software is still closed and expensive. With the opening of source code and other trends to make everything open, such as cloud computing and the consumerization of IT, the ICT world is changing rapidly.

I had the opportunity to moderate a panel session at this conference to investigate the current status of data center infrastructure management (DCIM). For details of this session, see here.

These were the panelists:

• Chuck Rego, Chief Architect, High Density Data Centers at Intel Corporation
  

• Pam Brigham, Director, Global Technology at Equinix

• Phil Reese, Research Computing Strategist at Stanford University

This blog is a summary of that session. (It is almost impossible to moderate a panel and take notes at the same time.) There are many ways you can structure a panel discussion. One extreme is for the moderator and panelists to share a common scenario, even down to the details of Q&A. Of course, the other extreme is to set a big theme and a direction for the discussion, and let the conversation take its own course.

I took the second approach, mainly because the panelists’ three data centers were drastically different, making it extremely difficult to ask each person the same question. On one end of the spectrum, Phil Reese has data centers for researchers at Stanford University and is starting to use a commercially available DCIM tool. On the other end, Pam Brigham's company, Equinix, is in the colocation business worldwide, and she uses homegrown tools. Chuck Rego produces a set of DCIM tools at Intel and uses other commercially available tools.

Technical difficulties prevented my monologue slides from being included in the presentation. But I said the following in them:

• DCIM tools are software and hardware tools used to design and operate data centers effectively. This definition may qualify almost any tool as a DCIM tool.

• In general, a tool has only one function.

• DCIM tools came out of the different needs and categories of data center operations. Therefore, there is no standard for sharing data and no common communications protocol.

• Very little information about use is available.

There were no clear disagreements about this explanation. However, Chuck was a little skeptical about any tool being a DCIM one. I am not 100% sure, but I think I heard that energy management tools were not DCIM tools. I take a very liberal stance on the definition of DCIM. If we take the meaning of DCIM literally, any DCIM tool should directly touch the infrastructure. Energy management tools may not deal with the infrastructure directly but they do indirectly. If we draw a line to define what is a DCIM tool and what is not, it would be too cumbersome. I suggest putting everything into this category.

There were a few more topics discussed, although I am sure I forgot others:

1. Homegrown tools were developed when no tools were commercially available.

2. A dashboard display that integrates several tools' results would be desirable.

3. Some kind of standards are necessary.

As for item #1, both Pam and Chuck said why they developed their own. Pam needed to provide some kind of automated way to let sales guys know what colocation space is available at which data center, with some detailed specifications. One such tool is web based and provides information instantly. When there is no tool commercially available, you need to develop your own. Pam said she had been looking into commercially available tools, but none of them satisfies her needs yet. A tool needs to be flexible and customizable because no two data centers are alike. A tool without any flexibility may apply to one data center but not to another, even though you own them both.

Chuck's case is interesting. He developed several tools as a suite to meet his needs but ended up making them commercially available. So Intel eats its own dog food.

I think both Phil and Chuck brought up item #2. Phil is using SynapSense to monitor his data center. He also has some CFD tools. Down the line, he will need more tools. It would be very desirable if these tools were integrated with one display window, rather than multiple windows, to make it easier to grasp what's happening at your data center.

Item #2 brings up item #3. To integrate tools together, we need a common platform for sharing data and a communications mechanism. But because each tool was developed to perform one function and one function only, this need was not taken into consideration. However, there is some movement in this direction. Future Facilities now teams up with other companies, including Intel, to integrate their tools together.

In summary, the DCIM segment is in its infancy. Its definition is not even agreed upon. There is going to be debate over whether a tool belongs to DCIM. That would confuse the market, but it is a process we need to go through to mature this segment. But one thing is clear. Someone with a lot of weight behind him should take the initiative to set the standards in this segment. Chuck, how about you?

Maricel Cerruti of Teladata

They are pretty funny and I want to introduce them to you.

Number 5:

Number 4:

Number 3:

Number 2:

Number 1:

They are not only funny but accurately describe the current disconnect between IT and facilities departments.

The way things are going, regardless of what we want, an extremely large amount of data will continue to be created, even at this moment. It seems the growth rate of data or Big Data will accelerate rather than slow. The term I learned from one of the presenters at the conference is data exhaust. The idea is like car exhaust: a lot of data created but not utilized, then discarded. If we can exploit this exhaust and mine good information out of it, it would be beneficial to business and energy consumption. Well, energy consumption at this point is not emphasized. It is information and communications technology (ICT) which generates this massive amount of Big Data. And it is ICT that exploits Big Data for our benefit. But ICT has been criticized by some because it consumes a lot of the nation’s energy. Therefore, I think it is also ICT that should control and conserve energy consumption.

Because the exploitation of Big Data has been discussed mostly from the business benefits perspective, let me focus on energy consumption. As I listened to several presentations, I wondered whether the following formula holds.

(Energy saved by having Big Data and analyzing it) > (Energy spent in creating, analyzing, and storing Big Data)

When gasoline was very cheap, large cars with big engines were manufactured and marketed. People liked them better than smaller cars with tiny engines. As time passed, the cost of gasoline went up and people became more conscious of air quality and the finiteness of energy. After the first Big Data hype is over, will we be in a situation similar to having more energy efficient cars? Or am I wrong not to factor convenience and other elements into the formula? Am I saying something like "We should not heat our homes, to conserve energy during a severe winter”? It does not make sense at all. There should be a reasonable balance between comfort and conservation.

To make sure that the formula holds (or at least try to do so), we can make the left side larger and the right side smaller. For the former, we need to increase the effectiveness of the data and/or information to tune up energy use. For the latter, we should develop methods to effectively perform these three functions. If we want to increase the effectiveness of data, we may want to collect it more often. But simply doing so will drastically increase the amount of data to be collected, analyzed, and stored. For instance, we could collect power consumption data on a particular server every millisecond, but it is very unlikely such frequent collection would increase the accuracy or effectiveness of the data. There should be some optimal interval for collection. Because monitoring and measuring are pretty new in many data centers, it is hard to come up with the optimal frequency of collection.

Furthermore, the collected data must be stored and analyzed, which at some locations will consume energy as well. The more data we store, the more energy is required. Analysis is the same way. The more data we have, the more energy is required to analyze it. Again, there should be an optimal level of data just to be energy efficient.

Maybe this entire argument does not matter. The magnitude of data we can control, as in the case of a particular data center’s energy consumption, is nothing compared with the magnitude of data generated everywhere in the world 24x365. That data is created very rapidly and accumulates in an exponential way. I feel so powerless when I think that even if we consciously try to control the creation and storage of data by ourselves, it probably won’t make a difference.

People are becoming very conscious of e-waste, electronics such as computers and display terminals that are outdated and no longer necessary. I wonder if we will have something like e-data waste or e-information waste soon. That is, the data that was collected and stored but no longer applicable or necessary. Because of its magnitude, no one really knows where it is or whether it is even wasted. With e-waste we at least know the equipment is no longer necessary and can take appropriate action, such as recycling. But with e-data or e-information waste, most of the time we do not know where it is or whether it even is waste. How do we know some data or information is waste or not? Some data and information become outdated and may be classified as waste. But some data and information, like that related to history and art, may not easily become outdated, even with the passage of time.

As technology progresses, it could become possible to capture all the data and information without deleting it. I acknowledge that I am a culprit myself or at least encourage the trend unintentionally. I use Gmail along with other free email services. Because Gmail allows me to keep all my emails, I seldom delete them, except for spam. My emails are stored on some disk or backup storage somewhere in one of Google’s data centers. Do I have an incentive to go over them to delete emails no longer necessary, such as those from several years ago? Well, quite frankly, I do not think so. So I am as guilty as anyone else who is enjoying this luxury.

I do not know whether e-data waste will become a problem in the future or how many more years it will be before it does. Maybe such a worry has no substance but I wonder….

Mr. Matsumoto passed the torch to Mr. Kimura (in the next picture), the new CEO of Fujitsu America Research Labs, who opened the conference.

Yasunori Kimura, new CEO of Fujitsu Laboratories of America.

The conference included very interesting information about Fujitsu itself and its take on Big Data. In addition, there were expert speakers in the Big Data segment. Fujitsu also exhibited 19 of their technologies at the conference. The agenda and other information are here.  As in previous years, most presentations will be posted here. 

There were several presentations from Fujitsu, but I show only the slides, with my comments, that I found most interesting. A complete set of slides will be sent to all the participants and will be posted here soon. 

In this slide, three basic research areas are shown with the core strategic R&D expanded. Cloud fusion and next-generation green data centers are the most directly related to Big Data. By the way, I do not see much use of the term cloud fusion by others, except for Fujitsu. Looking at their explanation (the 11th slide in here), it seems to be similar to cloud federation, which includes multiple private and public clouds. Such a federation will make it easier for enterprises to use multiple kinds of public and private clouds on-premise, or it may pose extra problems, such as identity management and interoperability of virtual machines (VM).

In the next slide, Fujitsu showed where data are coming from to form Big Data.

Fujitsu is serious about riding the Big Data wave. Their specific project areas are in the next slide.

I bumped into Dave Ohara, who has been writing a lot on data centers. He is now into Big Data, and his report on it will come out soon and be available at GigaOM’s website (for a fee).

Unless I make an energy-related comment, this blog is not complete. The current focus on Big Data is pretty much on how to exploit it. It is important to utilize "data exhaust,” similar to exhaust from cars, and make good use of it. Even in this instance, a large amount of data is created 24x7 in almost every corner of the world. Much of it is created via machine-to-machine (M2M) interactions. Some of it is discarded as exhaust, without any processing, while some is stored and accumulated with very little processing. Only a small percentage of it gets processed to produce useful information to improve our lives and accelerate smarter use of energy, as by smart grid.

Come to think of it, this is very scary. Many kinds of waste, including garbage, as long as they’re biodegradable, go back to nature. Of course, some do not. GHG disappears from our sight quickly but accumulates high up in the atmosphere and impacts the earth. But unused but accumulated Big Data will never leave us unless we delete it. It usually takes energy to collect and store Big Data. I wonder whether this will become a problem sometime soon. There are certain aspects of Big Data we can control, but we cannot control all of them. Am I worried about something that will never happen?

These days, $40 to $50 a month connects you to the Internet with almost no restrictions. I am one of the guys who are addicted to the Net. I am online most of my waking hours. When I try to recall how things were before the Internet, I cannot remember. Actually, I cannot even remember how things were before broadband access. I often make the 11-hour trip from San Francisco to Tokyo, yet I can converse with and see my relatives and friends in Japan in real time at my kitchen table in Silicon Valley. Email arrives there and everywhere else instantly, without any delay. I can access most websites, including video sites, regardless of where they are. Don’t you want to know how these things are possible? I do!

I found Bill Norton at an ISP conference some ten years ago and listened to his talk on Internet peering. From his session and casual conversations with him after that, I had some idea of the Internet’s inner workings but no details. I’ve followed his work off and on over the years. I am no expert in the details of the Internet, but I have a ton of interest in and curiosity about them.

After reading his book, The Internet Peering Playbook, published in July, I think I understand one aspect (peering, that is) of the inner workings better. Three things came to my mind as I read:

1. Is the Internet truly a free forum?

2. Is the energy efficiency of the Internet considered?

3. Can Bill reveal such secrets and stay in this segment?

As for the first point, I had believed that the Internet is a place where people can interact with others without any restrictions. Free flow of information is a founding principle of the Internet, yet the core of the Internet seems to be controlled by a handful of big ISPs (Tier 1) and intermediaries of content providers. Even if I find an interesting website via a search engine, I will not access it unless I can do so quickly, and that may not be possible for business and competitive reasons. Access may be too slow, beyond my patience, because packets needlessly travel an excessive distance. Maybe the Internet is generally, but not always, a free forum for exchanging data and information.

Regarding the second point, in considering how much energy ICT consumes, we tend to pay attention to servers and storage equipment because they are easily visible. Networking, on the other hand, usually does not get much attention. Computing is moving from in-house to clouds, many analysts and ICT experts say. The move towards clouds seems to be inevitable. You’ve heard the benefits of such a move. One is that cloud computing can realize green computing, mainly because of economies of scale. Several recent studies, such as one from Accenture, attest to this claim. "Not necessarily so,” say researchers at the University of Melbourne. 

Depending on the type of traffic between clients and clouds, the energy consumed in their communications may be much more than the energy saved with cloud computing.

If that is the case, my inquiring mind asks this question. If I peel off one layer of Internet communications and go down to the next layer, will I find that packets are routed from one corner of the world to another in the most efficient way, that is, by the shortest path? After reading Bill’s book, I know that is not always the case. Depending on how one ISP peers (or not) with another, a packet may have to travel extra miles, going abroad over the Pacific or the Atlantic Ocean and coming back to the destination ISP’s customer. The average customer cannot tell how each packet is delivered, much less how much energy is wasted on extra travel miles.

I understand that the Internet works and grows at a tremendous pace because the business models of those who participate in its infrastructure and operations work. And I think any controls on them would kill this good growth. Maybe it is too naive to hope that energy efficiency does not get killed by business growth. I wonder if there is a business model that makes money by paying attention to energy efficiency, such as that provided by packet routing.

For the third point, Bill revealed many tactics and tricks used by ISPs to gain their position and increase profits. Such information is only available to a select few. I’ve never seen anything like this anywhere else. I jokingly asked Bill if he would not be ELIMINATED after publishing this book.

Finally, let me just summarize his book briefly:

• Introduction

• Section I: Chapters 1 through 3 introduce Internet transit and explain what it is. This is important because transit and peering are two different things. This section also discusses the tactics used for transit.

• Section II: Chapters 4 through 7 discuss peering.

• Section III: Chapters 8 through 10 discuss the peering ecosystem.

• Section IV: Chapters 11 through 14 cover 27 tactics for approaching target peering partners and some data center details of networking.

Every chapter contains invaluable information, but I liked Chapter 10 (The 21st Century Internet Peering Ecosystem) best of all. This chapter describes how the major players play their cards to continue to control their influence on the Internet. With the change in the market, some incumbent players have lost out to other guys. And that is well explained in a peering window.

Finally, even if you are not responsible for peering for your ISP, involved in data center operations, or an expert in the Internet, if you are interested in how the Internet works, you will enjoy this book.

• Tools Needed to Manage Data Centers
• The New Data Center Infrastructure Management Segment
• Chatting with Sherman Ikemoto of Future Facilities

As most people in the data center market know, both facilities and IT folks consider monitoring one of the most important elements in operating data centers. Smaller companies were the first to provide monitoring and reporting functions. Although this is not an exhaustive list, I had a chance to talk to some of these vendors and write about the meetings:

• Visit to OSIsoft
• Visit to Sentilla 
• Visit to Modius 
• Visit to Arch Rock

I understand their services and their usefulness. Some provide sensor hardware and software, but others provide only software. They all monitor, aggregate, and report several parameters relevant to data center operations, such as temperature, humidity, and power consumption. Some deal only with facilities equipment, and others handle data coming from both facilities and IT equipment. There are no standards by which to measure the data—no standard for frequency of measurement, data formats, or protocols. Each vendor has their set of customers, and they seem to be happy with the solutions they purchased.

Then there are Power Assure, Romonet, and Future Facilities. Power Assure does monitor, but that is not all. It also optimizes the use of power at your data center. Romonet is for capacity planning. Future Facilities provides an electronic version of a data center that you can play with before implementing your design physically. These three cannot be classified as monitoring and reporting vendors. But their functions are important to operating data centers, in addition to monitoring and reporting, so a new term has been introduced to describe a new segment, which is DCIM.

Clearly, DCIM should contain several categories of tools, including those for monitoring and reporting, capacity planning, and simulation. As I said before, this segment is in its infancy; there are no standards or actual-use information. Those who combat day-to-day operation problems would be confused about which tools to select. Do they want to buy one tool at a time or buy a suite of tools? But wait. There is no suite of tools yet, although Future Facilities (for example) has begun to partner with other DCIM vendors to share data.

If we were to develop a suite of tools or a framework or platform for DCIM tools, what would the requirements be? It would help if there were some information from actual use by someone other than the vendors. Because DCIM tools are at a very early stage, there is very little information about them.

To fill that gap, I put together a panel discussion on DCIM at Teladata’s Technology Conversion Conference to be held on February 2 at the Santa Clara Convention Center.The program is here. 

Because the needs of operators can be quite different from one data center to another, we will have a good assortment of panelists from different environments:

• Chuck Rego, Chief Architect, High Density Data Centers at Intel Corporation
• Pam Brigham, Director, Global Technology at Equinix
• Phil Reese, Research Computing Strategist at Stanford University

Chuck develops Intel’s DCIM tools for their own and partner use and use commerical ones as well, while Pam at Equinix has homegrown tools. Phil at Stanford is starting to use a commercial tool. I will ask them what problems they perceive as the most important to solve at their data centers and why they chose their solutions, whether their own or commercial tools. Are they quite happy with the tools they are using? If not, what is missing? What additional work is needed to make them work? Conversely, were there any extra benefits they did not expect in applying their DCIM tools?

If you are interested in the answers to these questions, join me and the panelists at the panel and other sessions at the conference.

Greg Ness in the lobby of V3 Vantage Data Center.

Actually, this is not my first visit to their site. Back in November 2010, I attended a meeting there sponsored by Critical Facilities Roundtable. At that time, there were two buildings (V1 and V3) on campus that were owned by Intel and sold to Vantage just prior to the meeting. We had a meeting at V1 (well, actually before it became V1) and toured a partially completed V3 (yes, before it became V3). They added a brand-new building that is called V2, which was a large parking lot then. Right now, both V2 and V3 are ready for occupancy. V3 has three customers (one of them is Mozilla), and V2 is leased to a single unnamed customer. V1 is still undergoing renovation. The order of completion is V3, V2, V1.

All the specifications of each data center as well as their pictures can be found here:

• V1
• V2
• V3

But for busy readers, I summarized some basic information in the following table.

For more details, visit the respective pages. The total IT load of 37 MW is supported with a dedicated substation (50 MW) on campus.

They also are building a data center in Quincy, WA, to exploit abundant and cheap hydropower, thanks to the Columbia River, as the likes of Amazon, Microsoft, and Google did.

There are a lot of data centers in Santa Clara, mainly because of the:

• Reliable power (good mix of green power) with a competitive pricing structure from Silicon Valley Power Company, owned by the city of Santa Clara.
• Abundant fiber connections close to a NAP.
• Nearby large businesses like Intel and Yahoo (Google, Apple, HP, Cisco, and Oracle are not far from Santa Clara) and technology startups.
• Favorable weather for natural cooling, a.k.a. airside economizer.

I have visited data centers by Terremark, Digital Realty, and DuPont Fabros before, which are in the vicinity of the Vantage centers. They all cited the reasons above for selecting Santa Clara. They are alike in some ways and different in others.

Is it good to be in the data center business now, even in this time of recession? Absolutely! As you’ve probably heard from several sources, data centers are in high demand. At the first wave of the Internet, it was said that "if you build it, they will come.” At that time, "it” was fiber cable infrastructure. I will not be too far wrong if I apply the same quote to data centers.

Briefly, demand is high because:

• Many more things are online (cloud, mobile, sensors, and EV)—the Internet of things.
• Building and maintaining a data center is very expensive.
• Construction takes a relatively long time (12 to18 months), which delays go-to-market timing.
• Data centers require highly skilled experts to handle ever-increasing IT demands, but acquisition of such expertise is hard, time-consuming, and costly.

Vantage is on this wave.

The following is a summary of the visit and the discussion I had with Greg.

Several features differentiate Vantage from others. But one thing that stuck in my mind is LEED certification. They want to certify all four of their data centers as LEED Platinum. V3 has been awarded it already (see the sign next to Greg in the picture above), and V2 is prequalified as Platinum. Greg told me that they exceeded even the Platinum requirements, and the US Green Building Council (USGB) folks are considering incorporating the additional Vantage features into the certification.

At this time, there is no version of the LEED certification specifically for data centers, but it has become the de facto standard used in the data center market to brag on how green their data centers are. EPA has finally nailed down a version of Energy Star for data centers, which indicates a certain level of energy efficiency, but it has not caught on yet.

Speaking of catching on, PUE has caught on with almost everyone in the data center industry, but it has been misused to promote data centers too much. Power consumption by IT equipment can vary depending on where you measure—such places as UPS, PDU, or IT equipment. Also, PUE changes with the time of day as well as with the weather and the season. Many operators used to report their best PUE to claim how green they were. To correct this and standardize the way PUE is measured, Green Grid has issued a few papers on it. Most recently, they emphasized that PUE should be based on energy instead of power and that each PUE should be qualified depending on where power to IT equipment is measured. In my conversation with Greg, this new PUE was not mentioned. By the way, they measure the IT power consumption at the PDU, which is quite good because they do not have the means to get to their customers’ IT equipment. The PDUs are a demarcation between their responsibilities and their customers’. In this way, they can clearly see each responsibility and have a good basis for the PUE calculation.

As has been said in many places, PUE by itself should not be used as a marketing tool but rather to compare progress on your own data center energy efficiency. The PUE numbers for V3 and V2 are estimated to be between 1.10 and 1.30. V2’s number (which has not yet been announced) will be better primarily because it is a brand-new construction. You can design energy efficiency into a new data center from the ground up. In spite of heavy modifications, retrofitted buildings inherit the old shell and design and cannot be made to comply completely with your new energy efficient scheme.

Another differentiation is the reliability of power and cooling. With a few exceptions, they employed the 2N or mesh policy to guarantee reliability for such things as UPS and generators. A data center supports the nation’s information and trade infrastructure; it cannot fail and must have a high level of availability and reliability.

Greg, like Jim Trout, their CEO, said that Vantage comes close to Google, Microsoft, and Facebook in terms of energy efficiency. It is not fair to compare a wholesaler data center player like Vantage with them. It is like comparing the performance of an embedded system with a system that runs software developed independent of a hardware platform. Because you can control and tune both software and hardware in an embedded system, your performance is better than in a system consisting of independently built software and hardware. A data center operator cannot influence their tenants too much. However, in the case of their V2, Vantage worked closely with the single tenant, who signed a lease during construction, reflecting their IT requirements for energy efficiency. Of course, they could not dictate which IT equipment to use or its basic design, as Google and Facebook did to their internal IT group.

What about the downside of this? If the building was spec’d to a specific customer, it is great for that customer. But is it good for Vantage? What if the customer goes bankrupt or leaves for any other reason? Aren't they stuck with a data center too tailored to be used by other clients?

The answer to this question is twofold:

• If the building were designed to be super energy efficient, it would last 20 years rather than 5 to 10 years, which is typical of data centers in the past. No significant retrofitting would be required for the next customer.
  
• If the customer goes under, it is likely that their IT equipment would be left there, and that might attract other customers.

Data center infrastructure management (DCIM) tools are being talked about everywhere. For a wholesaler like Vantage, they only need to monitor and control the facilities side of the house and not IT. For that they use IBS.  Incidentally, I will moderate a panel session, and one of the panelists is from Equinix, a colocation provider that does not need to measure IT equipment either. I wonder what she will say about it. 

In V3, a raised floor with three separate caged areas is employed, instead of a slab floor. This allows customers to adjust airflow on the basis of their usage and configuration. However, the cold/hot isle configuration is pre-architected with the ceiling heat damper. So unless customers do not want to follow one of the best practices of energy efficiency, their rack configurations cannot be drastically different to exploit this preconfiguration. Each customer is allocated a single PDU, which allows up to 3 MW of power. As for cooling, both V2 and V3 employ airside economizer. In addition, V1 and V3 use chiller for cooling, but V2 does not. The customer in V2 does not want to use water and uses DX instead.

It is sometimes more profitable to conduct business at a higher level. So rather than competing at the infrastructure level, Vantage may want to provide a service like hosting private clouds. But I was told that they would strictly stay on their core of competency and the area of their strength, which is providing the best efficient mechanical and power support. Something like hosting private cloud is left to their alliance partners, like Telx. Incidentally, on this tour, I bumped into an old friend of mine who works at Telx and will try to get time with him to discuss their services.

What I liked most is that Greg wanted to make some of their ways of energy efficiency available by showing how they do it, so that the entire data center industry gets improved. The way things are going, more demand for data centers is expected, and this market is one of the rare segments that do not know what the recession means.

He recently wrote a book, The Internet Peering Playbook: Connecting to the Core of the Internet, which you can find here.

I plan to review it. In addition to a normal review, I plan to consider whether peering has anything to do with energy saving and efficiency, because traffic on the Internet is expected to skyrocket, and each and every operation, including peering, will have some impact on how energy resources like electricity are used or wasted. So stay tuned.

Here’s a short summary of the workshop.

Like similar organizations in the United States, JDCC has several special interest groups, which they call working groups. Different groups focus on planning, market research, environmental policies, human resources, international competitiveness, environment and standards, facilities standards, security, and networking.

The first presentation was by Atsushi Yamanaka of IDC Frontier, a data center operator, who reported on power saving efforts in the Tokyo region. Such information will be useful for data center operators in the Osaka region because that area will probably experience the same hardship this winter that Tokyo did this summer.

Recall that Yamanaka and I presented the state of data centers in Japan in the aftermath of the major quake in March, as reported.

After the JDCC organizational introduction, Hiromasa Fuchi of NEC explained what NEC did to save power. He also referred to what JDCC published to help save power for its members.

Following that, Kohsei Ichikawa of Kajima (the largest construction company in Japan) described some details of the March earthquake. He said that one high-rise (53 stories) building in downtown Tokyo shook continuously for 13 minutes, even though the actual earthquake lasted only 60 seconds or so. He also reviewed JDCC’s new requirements for earthquakes. Among the new findings in the aftermath of the disaster are that the power grid can go offline (its availability used to be as natural as air before the quake) and that in spite of preferred customer contracts to obtain fuel for emergency generators, the breakdown in the supply channel prevented the securing of sufficient fuel.

Fukuda of Mitsubishi Research Institute briefed us on what the Tokyo Metropolitan government is doing about GHG control as it relates to data center operations. Osaka does not have a similar law, and data center operators in Osaka could learn from this example.　

Haruyoshi Yagi of Fujitsu talked about the environmental standards working group’s efforts to improve PUE, then Naohiro Masunaga of Nomura Research Institute discussed his working group’s findings after visiting data centers in Singapore and Malaysia.

Ogura of Hitachi reported what is needed to secure good people for data center operations in Japan. I then gave a talk on data center trends in the United States. My topics included DCIM tools, higher temperature in cooling, modularity, metrics, hardware/data center standards by the Open Compute Project, electricity distribution, and the impacts of cloud computing.

Sherman Ikemoto

At that time, Future Facilities’ (FF) main focus was computational fluid dynamics (CFD), which was important then and still is today. But it was not interesting enough for me to write about it (sorry, Sherman). In 2011 FF came out with new positioning and a new set of functions, a virtual facility (with a suite of tools called 6SigmaDC), a digital replica of a real data center. The virtual facility can put together information on power, IT loads, and space, in addition to air flow, and create a mathematical model and run simulations on it without actually altering a data center.

I had an opportunity to listen to people who are using this product at the recent FF conference. As I listened to their talks and had a frank chat with Sherman, I began to think that this replica has good potential to solve a big problem of IT and facilities: disarray in managing high-power-density data centers.

This blog is a summary of my chat with Sherman and my thoughts triggered by it. FF did start its business with a focus on air flow (the term DCIM did not exist then anyway, although CFD is one of the DCIM categories). He said that earlier they were brought in by data center facilities folks to clean up the damage done by IT. The use of the word "damage” was interesting because as a former long-time IT guy, I never thought facilities people felt that way. Facilities people tailor air flow to IT needs at the beginning of IT deployment. But because the IT way is notoriously to change everything—including equipment, rack configurations, and rack layouts—often and on-the-fly, air flow customized before the changes no longer applies after the changes, and thus IT does damage to operations in the entire data center.

After seeing this repeated again and again, Sherman and his folks realized it would be better to let IT and facilities folks work together to share air flow and other information to avoid the problem early on rather than fight with it later. Earlier in the conference, Hassan Moezzi, director of FF, said that air flow is the single most important factor in managing a data center because most data centers are cooled by air rather than liquid (such as water). By controlling air flow and optimizing its effect on cooling, most problems could be solved.

I think I knew this, but until it was put that way I did not fully appreciate it. Another thing I re-realized concerns IT and facilities integration. Since the beginning of my writing about the data center segment, many people have said that the difficulty of managing data centers is primarily IT and facilities’ differences in culture and lack of close collaboration. Some remedies were suggested, such as making both IT and facilities report to the same boss and/or letting IT be responsible for the power bill. Those are fine, but they are at too high a level. What can we actually do? Sherman and FF are advocating to create a digital replica (mathematical model) of a physical data center. The model is used to test multiple data center configurations to find the best before putting the real IT infrastructure in place. This makes sense. I have toured many newly constructed data centers. Standing in an empty floor, I often wondered how they would lay out IT equipment to manage the entire data center in an energy efficient way. They do not know in advance how the IT equipment will be laid out and how electric and mechanical systems can support it. Come to think of it, it is a scary thing.

Now my next questions. Developing a mathematical model is fine, if we are talking about new construction. Granted that many new data centers are popping up everywhere, including Silicon Valley, there are a far greater number of existing data centers. If the model cannot apply to existing ones, FF’s solution is very limited. But if it can, that means a great business opportunity. FF is often called in to find a solution for an existing data center that has extra capacity (in theory) to host more IT equipment but cannot expand further for some reason—maybe there are hot spots. This is called stranded capacity. By diagnosing the root cause, they can fix the problem by constructing a virtual facility and analyzing it.

This is great, but there is no mathematical model for existing data centers, which consist of hundreds and thousands of pieces of IT and facilities equipment. How do you collect a list of equipment and logical connections to construct a model for that? Initially, FF collected and entered information by hand, a time-consuming and error-prone process. Later, they created an interface to bring in data automatically from multiple sources, such as IT configuration databases that might be produced by someone like Asset Point with their autoscanning of IT equipment. With this interface, FF could work with a company like Nlyte.

A natural question is whether there exists a standard for a communications protocol and data format to share the data created by each DCIM tool. Unfortunately, at this point there is none, although FF uses XML as a base. Even with XML, you could still have your own data formats, although it might be easier for conversion because XML is ASCII based. In any event, FF developed their own interface and data formats, which they share with their partners, like Intel, Nlyte, Aperture, RF Code, and SynapSense. This allows assets and monitoring information into the virtual facility model.

Well, this is interesting. It would be great if FF, or whoever leads the standardization of data formats, could integrate many more DCIM tools into their virtual facility platform and accelerate the adoption of DCIM. I explored this in my previous blog. 

FF is working with Intel as a development partner, and their solution interacts with Intel’s data center manager (DCM). Intel has established an interface with data coming from servers and is working with FF to merge their interface with it. Since the DCIM market is in its infancy, there are no standards. Cooling and electrical solution providers like Schneider and Liebert-Emerson and others have their own interface and data formats. I know Intel is big and that more than 80% of all the servers in data centers run Intel chips. Is Intel powerful enough to force a standard to unite DCIM tools? After all, we need to convince facilities types to agree on a standard, and they are not used to standards.

Sherman thinks that the most important thing for really optimizing the efficiency of data centers is to understand data from servers, which is the real culprit, not cooling or electrical systems. "If Intel controls such data, why not?” he continued. It would be IT, not facilities, that would set the standard, he said.

This argument is convincing, but my skeptical nature forces me to wonder if the facilities type would go for a standard. In the BMS market, vendors were forced to support an interface with the Web because the Web revolution was so powerful that they needed to support the Web/IP protocol. We need a similar magnitude of scale to force the standardization of data formats so that each DCIM tool can share information on a single platform like FF’s. I do not have any idea what that would be. Would it be a power crunch, I wonder?

How about adoption? FF has roughly two types of customers: Web/Internet and mission critical. The former includes Intel, Facebook, Google, and Microsoft. The latter includes Bank of America, which will soon announce its adoption of FF’s solution, and JPMorgan Chase. FF is also targeting medium-size data centers, as they expect them to get the same benefits as large data center players. The company originally came from Europe, and their presence there is fine. But they have yet to penetrate the Asian market, although they have customers there for designing server boxes with their tools.

As for channels and reselling their products and services, EYP/HP might be the closest to being certified, as FF is in discussions with them.

As Chuck Rego of Intel mentioned to me, we need to cover both the monitoring and the capacity planning sides of DCIM. If somehow FF can standardize the data for DCIM and unite both sides, DCIM will make it mainstream, and many of the "damages caused by IT” may be avoided.

SVLG stands for Silicon Valley Leadership Group, whose vision (from its website) is to:

Ensure the economic health and a high quality of life in Silicon Valley for our entire community by advocating for adequate affordable housing, comprehensive regional transportation, reliable energy, a quality K-12 and higher education system and prepared workforce, a sustainable environment, and business and tax policies that keep California and Silicon Valley competitive.

SVLG has been offering data center energy efficiency (DCEE) conferences for the past three years. It is situated in southern San Jose and surrounded by a large, natural open space.

Like the previous conferences, this one was sold out, and a lot of people showed up.

Crowds at SVLG DCEE conference. There are more rows behind this.

The agenda for the program was as follows:

I will report on the entire conference, and some sessions in detail, in upcoming blogs.

Carl Guardino, CEO of SVLG, gave the opening remarks. Although he covered a few topics, I only remember one thing (sorry, Carl!!).

Carl Guardino

It is a joint program between SVLG and the City of San Jose to donate your time and money to match 1,000 adult volunteers with 1,000 struggling K–8 public school students to help eliminate the achievement gap in SV.

More can be found here.  One type of contribution is to spread the word. So this is a part of my contribution.

Stranded capacity is capacity that IT cannot use because of a data center’s configurations and layouts. When we design a data center, we prepare enough power and cooling to meet IT needs. But depending on how we deploy facilities and IT equipment, we may not be able to fully utilize the capacity allocated. Chuck started his talk by saying that he wanted to get a handle on stranded capacity by measuring and quantifying it.

The two most important things in data center operations are reliability and utilization. When we discuss data center energy efficiency, the most-used metric is PUE. Chuck was one of the original five people who discussed the definition of PUE, even before The Green Grid (TGG) defined it officially. What is missing from the current PUE is the incorporation of load information. TGG has been working to incorporate IT utilization and other information to improve PUE. Another metric, CADE, which was suggested jointly by Uptime Institute and McKinsey, also considers the utilization of IT and facilities equipment in its definition. However, I am afraid it has not caught on with the majority of data center operators. PUE is still the dominant metric for energy efficiency for data centers.

Chuck wanted to find out how utilization information might have an impact on PUE. He set up a model that assumes an average of 8 kW/rack and a peak of 12 kW/rack. With this assumption, we can obtain fairly low PUE. Does this level of PUE hold when the pattern of operations changes? What if we calculate PUE for an environment where IT utilization is low? With this average and peak power requirements assumption for a data center, PUE is 2.0. But under a utilization factor of only 20%, the actual operating PUE goes up to 5.7. This is because other supporting elements (both mechanical and electrical) were set up to support much higher loads. He calls this type of PUE actual operating PUE. The point is that the way you operate your data center could make a big impact on the actual efficiency of your data center, even though it was designed to be energy efficient for average utilization.

Hassan Moezzi, director of FF, said that there is a disconnect between the operations of the entire data center on the one hand and server design and rack configuration and layouts on the other. Most IT folks, including me, do not know or care how each server is built; we’re not going to open up a chassis and carefully review the components. According to Chuck, factors like the following may make a 10% difference in energy efficiency:

• Shadowed or unshadowed processors (relative positions of multiple CPUs have impacts on the cooling efficiency of each CPU)

• Processor efficiency based on different levels of workloads

• Fan speed control

• Heat sinks

Even at 21°C, these affect efficiency, and under ASHRAE’s increased temperature and humidity setting of 27°C, the difference would be much more.

Chuck conducted experiments to find out what impacts air flow has over data center operations. He learned two things from his experiments. One is the importance of finding the optimal location to measure temperature. Traditionally, it is measured at the return points of each CRAC unit. His experimentation indicated that temperature control should be done at the supply points (inlets to servers) rather than at the air return points at CRACs. At the return points, there could be some complex air flow, so they may not accurately reflect necessary cooling requirements for server loads. In his experiments, the temperature oscillated widely at the return points, while the supply temperature stayed pretty much constant. As the temperature is increased from 21°C to 27°C, this trend would be amplified.

Another finding was the need to set cooling at a higher temperature. At higher temperature, cooling needs are relaxed, while the IT side may increase power consumption with higher fan speed and silicon leakage (at a higher temperature, CPUs tend to consume more power). So the difference between the gain by facilities and the loss by IT should be carefully weighed. In raising the temperature, reliability and performance should not be compromised. The experiment involved 900 servers for 10 months and tried several temperatures, ranging between 21°C and 35°C. But he did not observe any performance degradation or visible failures at all. This is quite impressive, with real data to back up the result.

Chuck then talked about the placement of sensors. If we want to obtain useful data from each server, we need to attach a sensor to each server. In a big data center, the number of servers can be in the tens of thousands, and it is not reasonable to assume we can attach one sensor to each server. He then talked about smart servers, which come with an embedded sensor. The measurement of relevant information, such as temperature, can be done underneath the OS (so that it is applicable to either Linux or Windows).

Moreover, cooling traditionally has been static and unchanging, even with different loads. But loads change dynamically, and cooling needs should change accordingly. Otherwise, some cooling capacity is wasted. When IT decides to move virtual machines (VM) from one server to another, the loading factor of each server changes with the changed cooling requirements. Power and cooling requirements also could be adjusted, if more accurate loading and operating data are available. Intel has a prototype to give feedback to the dynamically changing server environment and let some servers sleep to optimize the energy efficiency of the whole data center. The last time I talked to PowerAssure, their product had such a feature and worked with Intel.

Sherman Ikemoto of FF said that what ultimately decides energy efficiency for a data center is data from servers but not from facilities equipment. I was somewhat skeptical about that. But after Chuck’s presentation, I am more convinced of his opinion. Maybe we have been tackling the symptoms of the problem rather than its root cause. The problem is, in Sherman’s phrase, "damage done by IT.” But we were not dealing with the real problem of controlling IT equipment. Some time ago, Emerson issued a white paper on Energy Logic and claimed this about the power saving at the server level:

1 watt savings at the server-component level creates a reduction in facility energy consumption of approximately 2.84 watts

Although it was saying the same thing, it was not positioned to emphasize both Sherman and Chuck’s points. By changing the mindset, we may make progress in improving data center energy efficiency.

Before I started covering the intersection of ICT and energy, I was a diehard IT guy. My specialty was software engineering (do you know what that is?), which is a discipline for making software development a science rather than an art. In my humble opinion, this discipline can make the average Joe a reasonably good software engineer, but it cannot make him a master. The merits of software engineering aside, there is something called the software development environment or integrated software development tools, or some variation of that. When we develop software, we go through requirements, specifications, high-level design, low-level design, coding, testing, and maintenance. For each stage, we need a software tool. A simple example is Unix (no, not Linux; it was Unix in those days). Simple utilities came with the Unix operating system, such things as ed, sort, head, and cat. Those tools were well integrated with their environment (the Unix platform), and they worked in harmony.

As software became larger and more complex, more sophisticated software tools started to emerge. At the beginning, it was fine to have a single tool for each task, such as designing, coding, and testing. But people got more sophisticated and realized that software development could be done more effectively using a comprehensive set of tools that share the same environment.

I am not sure if the software development tools analogy can be used for DCIM tools. If it can, DCIM is going through the same cycle. Early DCIM tools were not integrated with others but independent of each other. They included monitoring (temperature, pressure, humidity, and power consumption), alarming, aggregation (both IT and facilities data), analysis, and display. More functions are added constantly. Kevin identified five functions in his second blog.

"So that makes five for DCIM: Inventory, Change, Capacity, Simulation, and Efficiency Modeling.”

I agree with his assessment. I am setting up a panel session for a conference next year to talk about the users’ perspective, not the analysts’ or vendors’, on what they need. Because the subject is so big, I will not be able to cover the entire subject. I will start with what a data center operator needs to run his own data center. What problems keep him awake at night? I would like to assemble a few data center operators that have adopted some DCIM tools, and hear from their actual experience with them. This might give us some clues to which tools are still necessary and how those tools should interact with each other. When I interviewed Modius, a DCIM vendor, their approach was similar to the integrated software development environment. I think that may be one way to tackle the issue. (I want to make it clear that I have no business or financial ties to Modius. I simply use that company as an example.)

Those who are interested in their architecture can see my blog on them.

In my recent visit to Japan, I discovered an interesting application currently under research by NTT DoCoMo, the largest mobile phone network provider in Japan.

It is called mobile spatial statistics, and is described here. In developed countries like the US and Japan, most people carry a cell phone when they go out. To stay operable, each cell phone is tracked by the closest cell tower so that it can remain connected to the wireless mobile network. A mobile phone network provider like NTT DoCoMo can identify each cell phone with its owner’s information, such as name, gender, age, home address, and current location. It is an invasion of privacy if such information is collected and used. However, if we simply detach this information from each individual identity, it is a valuable aid to providing necessary services and data for designing and altering social infrastructures and the distribution of goods.

DoCoMo provides two examples.

The first is that when the population of a city fluctuates, not all parts of the city lose or gain population equally. Currently, the only way to grasp the real movement of the population is to conduct a census. Even then, we can only understand the population spread statically, at people’s homes. In a typical day, people usually leave home in the morning for work or school, take a break for lunch, and go home after work or school. In other words, their movement is dynamic rather than static.

With mobile spatial tracking, we could track the movement of people in a semi-real-time fashion and use that information for city design. If some parts of a city have a denser population at a specific time of the day, we might add special services to alleviate the concentration. For example, if it is too congested at certain bus stops at certain times, we might add more bus services or encourage different commute hours to alleviate the congestion. If a library becomes very crowded at certain hours, we could shift its opening and closing times to solve the congestion. If we can see how people actually move around, we can take the necessary actions to accommodate their movement. This was not possible before mobile spatial statistics (MSS) because we could not obtain such information in semi–real time before. Even the census is taken only every ten years, which makes it useless for MSS.

The second example is to provide appropriate helping hands in the case of disaster. Although Tokyo did not suffer major damage from the March earthquake and its aftershocks in the eastern part of Japan, all its public transportation, including trains, was stopped for an extended period. Because of this, many commuters congregated at and around train stations. As the transportation shutdown continued, food, water, shelter, communications, and other human necessities were running low. In hindsight, appropriate help was not rendered.

Although efforts were made to ease the situation, there was no precise way to tell how many people were at each station, how many of them were elderly, how many might have medical conditions to take care of, and so on. If such information had been available, more appropriate actions could have been taken.

I can think of many more applications. MSS makes needs visible in semi–real time. If we know the details of the needs, we can provide the necessary services and goods to accommodate them. Needs may change with time and location. If we know what they are at a certain location at a certain time, we can send appropriate supplies. Without MSS, we may waste supplies or fail to meet needs. MSS can be thought of as energy efficiency. ICT can be linked to MSS to find an optimal way to provide supplies to satisfy needs.

MSS has a lot of potential and can be applied in any developed country.

The first speaker was Spas Lazarov of GE Energy Service, who spoke about points of power loss in data centers.

Spas Lazarov

The power use ratio in a data center varies from one data source to another, but Spas showed the ratio in the following picture.

Power use ratio in a data center

In his graph, I think the cooling ratio is a little too small, but it is within a reasonable range. The two items that consume the most power are IT and cooling. The power required for power distribution is 10–20%, small enough not to be very noticeable. There has been a lot of discussion about conserving power for IT (refreshing hardware and virtualization) and cooling (hot and cold aisle, containment, and air economizer). Although some work has been done, efforts to curb power consumption for power distribution have been minimum. Once power enters a data center, it changes its voltage and form a few times before it hits IT equipment. The following is a picture that shows the normal course of power distribution in a data center.

In a normal data center, UPSs are placed to prepare for power loss or poor quality power.

There are several UPS configurations, but Spas talked about double conversion mode and high efficiency mode. In the double conversion model, power, which entered the data center from the grid, always goes through the UPS. This is to ensure that IT equipment always receives good quality power in terms of frequency and voltage. Power from the power grid is AC. When that AC enters a UPS, AC is transformed via a rectifier to DC power for the battery inside the UPS. Then DC power is taken out of the battery and converted to AC power via an inverter. Two conversions introduce power losses, with an additional loss at the battery. If instead the UPS is operated under the high efficiency model, which bypasses the UPS in a normal situation and taps the UPS battery only in the case of an emergency, power loss is minimized, as shown in the next picture. In the figure, PUE increases because more power is fed to IT equipment rather than wasted in power distribution.

Spas pointed out that power quality in the US and most developed countries is good enough not to worry about. He gave us the example of a 50,000 square foot data center with a 5 MW load and a power price of $0.01 per kilowatt hour saving $3M annually by means of the high efficiency model. He also said that if all UPSs were operated under this model, $3B could be saved globally.

This is great, but what if the supply power is cut off from the grid? Can your servers survive the power interruption? Spas showed a diagram, below, with some details about the way power components are put together inside a server.

Normally, a server contains a capacitor. As in the power grid, which has capacitor banks at strategic locations like substations and distribution networks, this capacitor inside the server maintains power for the server in case of power loss from outside. As shown in the picture above, a typical capacitor can maintain power for 20 ms in case of power loss. When you check your server’s specification, that information should be described on it.

Spas also showed us that operating UPSs in the high efficiency model does fit within the ITI (CBEMA) Curve set by the Information Technology Industry Council. ITI is the high tech sector's advocacy organization in Washington, D.C., and in foreign capitals around the world. It deals with such issues as corporate tax reform, trade, telecommunications, cybersecurity, energy efficiency, workforce and STEM initiatives, regulatory compliance, accessibility, and environmental sustainability.

The ITI (CBEMA) Curve and application note describe an AC input voltage envelope that typically can be tolerated (no interruption in function) by most information technology equipment.

As I dive into smart grid and power engineering, I revisit the issues in electrical and mechanical functions in data centers. Although I came from the IT sector, I think I am beginning to understand the facilities perspective a little bit better.

Computing is changing rapidly from fixed location to mobile and from a large form-factor desktop to laptop and handheld devices. GigaOM threw a timely conference to focus on mobile computing. In conjunction with this conference, GigaOM published a report, The Future of Mobile: A Segment Analysis by GigaOM Pro, which comprehensively covers the field of mobile computing, provides some basic information, analyzes each major area and predicts its future, and lists players in each segment. The report is a densely packed 111 pages, but when you finish it, you will have a good picture of the current state of the mobile field. You can get it from here.

If I have any complaint with the report, it’s that it did not discuss mobile and energy. Well, maybe that is not a fair accusation because a discussion of mobile computing usually does not include its use for energy management. However, because mobile devices are replacing fixed and semimobile devices for computing, covering mobile and energy might have been a good idea.

The following is my description of the contents of the report. I don’t go into the details, simply because this blog couldn’t cover every subject with equal emphasis. Instead, I will mention some highlights that attracted my attention.

The report contains an introduction and twelve main subjects. I substitute my own headlines for theirs for brevity’s sake. Each section is written by a different author, and some by two authors. If you are overwhelmed by the report’s length, read the first two sections then skip to the areas you’re most interested in.

These are the sections:

1. Introduction

2. Platforms (OS)

3. Tablets

4. Mobile services with cloud

5. Connected devices (Internet of things)

6. Health care

7. Social-location-mobile and e-commerce

8. Mobile search

9. Hyperlocal targeting (digital coupon)

10. Mobile payment

11. Mobile advertising

12. Mobile VoIP and chat

13. Wireless mobile backhaul

1. Introduction: This is a good summary of the report.

2. Platforms: Platforms influence everything, and this section looks at the history of mobile platforms and predicts their future, with some advice from the lessons learned. One such lesson is to avoid channel conflict. As the battle between the iPhone and the Android intensifies, what will Google do to avoid channel conflict? Although other companies like Microsoft are not out of the competition, the two top dogs are the iPhone and the Android. This section is full of good lessons.

3. Tablets: The iPad has a commanding lead at this point, despite other offerings. Apple has set $500 as the price of a standard entry-level model. The price could fall, as more penetration is expected, or rise, with more functionality. The iPad will continue to dominate until 2013 or so, but by 2014 the number of applications available on the Android platform will allow Android-based tablets to overtake it. Amazon’s new version of Kindle may have its place in this segment. Finally, tablets will enter the enterprise market, first in vertical markets like hospitals.

4. Mobile services with cloud: Mobile services are enabled by applications on mobile platforms, derive data in clouds, and convert the data to useful information for consumption. Apple has an early lead in mobile services, but startups and established companies, such as Google, Microsoft, and Yahoo, cannot be ignored. Apple's new iCloud this fall is limited to Apple's platform but may support others. Potential M&As are likely in this space.

5. Connected devices (Internet of things): In 2020, there will be from 50 billion to 1 trillion connected devices, either single- or multipurpose. The latter includes smartphones, which were mentioned several places in the report as key drivers for accelerated mobile market expansion. To collect, process, and turn data into useful information, a set of tools will be required, including hardware, networking, and management platforms that will be hosted in the cloud.

6. Health care: Mobile health care uses wireless devices to manage health conditions, collect health data, monitor vital signs, provide clinical decision support, and access health information. Although now in its infancy, the mobile health market will take off globally, not just in the US, partly because of aging populations in many parts of the world. This space also includes wellness in addition to dealing with the sick, expanding its market applicability. For other drivers, read this section.

7. Social-location-mobile and e-commerce: Location-based services have been mentioned elsewhere, but no significant progress has been made. However, the confluence of social, location, and mobile, or SoLoMo, works well to connect consumers and businesses, providing what consumers want in their current context, including access to social media and their current location. SoLoMo with e-commerce opens up new markets.

8. Mobile search: Google and Microsoft have dominated search, but search in the mobile context is quite different from search on PCs and laptops. People with a mobile device are more location aware and demand more precise information about their current context. A keyword search based on PageRank generates more information than a small display on a mobile device can handle. New entrants are quite possible as the paradigm of search changes in the mobile computing.. One area that stood out and caught my attention is the virtual assistant. As with a human assistant, you do not need to specify the details to get your job done. The virtual assistant executes your job by automatically obtaining the necessary information.

9. Hyperlocal targeting (digital coupon): In short, this is a digital coupon. Consumers are more engaged when they are given coupons (perceived as content) rather than straight marketing messages. People with a mobile device are mobile, and their context, such as a location, changes constantly. Timely coupons and daily deals (with an upfront purchase requirement) will catch their attention.

10. Mobile payment: I was involved with a company that provided security to mobile platforms for mobile payment too early in that market’s maturation process. Now it seems that the market has finally started to move. For mobile payment to work, a good ecosystem needs to be built, including handset OEMs, software vendors, carriers, banks, and credit card companies. Near-field communication (NFC) has emerged as one of a mobile payment solutions and is expected to grow. The total global transaction volume processed through mobile payments will exceed $753.3 billion by 2015, rising from $31.8 billion in 2011. Mobile NFC transactions will be the primary driver of mobile payments through 2015 and will grow from almost $4.3 billion to $595.7 billion in 2015.

11. Mobile advertising: Spending on advertising in the US is $33 billion, but spending in the mobile space will be approximately $4 billion worldwide this year. This space will grow as the population with smartphones grows, because smartphone owners provide a rich target: They browse, search, consume content, and conduct commerce in and through their phones. Advertisers need to be there at those critical moments of purchase decision-making, which now takes place on smartphones—and to a lesser yet still important extent on tablets. Be sure to read about the seven trends in mobile advertising.

12. Mobile VoIP and chat: Juniper Research recently predicted that the total number of mobile VoIP minutes would jump from 15 billion in 2010 to 470.6 billion by 2015, thanks to the proliferation of 3G and 4G networks. The biggest boost will come from the US, which will account for 135 billion mobile VoIP minutes by 2015. It is somewhat difficult to make a precise prediction, because these features eat mobile phone companies’ lunch. One of the drivers is again a smartphone.

13. Wireless mobile backhaul: With the proliferation of smartphones with access to the Internet, the backend requires more bandwidth. Over the next few years, we expect to see increased adoption of wireless backhaul worldwide, with PMP and e-band technologies delivering increasingly attractive and cost-effective solutions for the new LTE networks. The companies to watch in this space are drastically different from the other companies to watch in this report.

The emergence of smartphones like the iPhone that allow access to the Internet and the proliferation of SNS make the mobile space quite different from legacy computing, and even from semimobile laptops. People are more context aware and demand data and information specific to their situation. Also, access to a vast amount of information and data from their SNS enables them to look for specific information. This is a game changer and there is no wonder that both Google and Apple, the two leaders in the smartphone space, show up in each category of companies to watch.

Again, the only complaint I have is the lack of coverage of mobile applications for energy. What was done at a fixed location for energy can now be done with a mobile device. Mobile should open up a new territory to monitor, control, and optimize energy in general. For example, a mobile device can be used to monitor and control energy use at home, for example, by following electricity use and remotely turning equipment on and off. Beyond the home, energy use can be monitored at buildings and factories. With appropriate data, utilities can dispatch repairmen to the exact location of a disturbance in transmission or in a distribution grid. More mobile applications can be developed to optimize the use of energy.

Finally, the report is well written and contains comprehensive information on the status of mobile computing. I highly recommend reading it.

For example, power usage effectiveness (PUE) takes only two points: total power consumption of a data center and IT power consumption. PUE has been used widely because it is simple and intuitive, and therefore easy to calculate and apply in any data center. Many other metrics have been proposed by adding other points, such as IT equipment efficiency. Unfortunately, up to now, no new metrics have replaced PUE. This does not mean that two points are good enough for measuring data center energy efficiency, but it does indicate how difficult it is to develop and improve a metric.

What is information and communications technology (ICT) sustainability? It consists of two things: sustainability of ICT itself and sustainability accomplished with ICT. IDC defined its own ICT sustainability index, applicable to each of the G20 countries, and defines it for the latter. High-level information is given in the press release, and details are given here.

Using a different methodology, Fujitsu, in its recently published global report on trends in ICT sustainability, defines its own ICT sustainability index (ITSx). Don't confuse it with the one from IDC; they are different. The report, ICT Sustainability: The Global Benchmark 2011, is available here. It is free, but registration is required.

Fujitsu’s measurement methodology was developed jointly by Connection Research and RMIT University. You can read the details in the methodology section of the report (page 40). In short, they identified four areas: (1) equipment life-cycle procurement and disposal, (2) end-user computing, (3) the enterprise and data centers, and (4) IT as a low-carbon enabler. Then for each area they created subgroups for attitude, policy, practice, technology, and metrics. On the basis of the answers to 80 questions from 1,000 CIOs and senior IT managers from seven countries, they scored each item in each subgroup and weighed them to produce an overall index.

I do not intend to summarize the contents. For those of you who want an abbreviated summary, you can see here. I only share some of the points that caught my attention in the report:

1. ITSx decreased in 2011 from 2010. The report speculates that it is probably because all the low-hanging fruit has been picked already. It looks like the next step is make ICT power consumption more visible to ICT managers

2. Visibility of IT power consumption increases ITSx. This is simple logic. If you know how much power your IT gear consumes, you tend to pay more attention to power consumption. The motivation here may be expense rather than sustainability. Nevertheless, it is a good thing and will increase ITSx for sure.

3. Canada has the highest ITSx, followed by the UK and USA. China and India were below average. No questions here. Both countries are busy with expansion and do not have much time to consider sustainability.

4. ICT/Communications/Media is the best performing sector. ICT is often to blame for its high power consumption, but it does its fair share to make other things more sustainable as well as making itself more sustainable.

5. Metrics and measuring are still poor. We definitely need to have easy-to-use and effective tools for this area. I have gone to many data center conferences where almost every speaker emphasized the value of measurement, but it seems that the message has not sunk in for the majority of data center operators.

If we have picked all the low-hanging fruit, the next thing that might increase ITSx significantly is measurement. This may not be that easy. We need to know what to measure and how often to measure. And what do we do with the data collected by measurement? Maybe the first thing to do is to let IT managers be responsible for their power consumption and get a bill for it. I think once they are shown the bill and understand that they are responsible for it, they will come up with many ways to slash it. That will stimulate more sustainability ideas because it will no longer be just research but reality to them.

Here is an aerial view of the data center:

This site is so familiar to me. I commute on Reed Street (the road at the bottom in the picture) every morning and evening to see how the construction is going. I sometimes walked from my office to the site to take pictures of the progress of construction. Over time (from 2008 to now), I came to feel like a father watching his kid grow. The dark cylinder shape in the picture is a thermal storage unit used for chillers. It can contain 500,000 gallons of water and is for phase 1. Phase 2 will build another of the same size. This picture was taken at least one month ago, and the cylinder is now painted a much lighter color. Just behind the thermal storage, there is a train track. I thought it was close, but it felt very close indeed when I was inside.

DuPont Fabros has published a good brochure describing the data center. You can take a look at the brochure, but briefly, gross space is 360,000 square feet and power capacity is 36.4 MW (18.2 MW in each of two phases). 

The presentation was given by Hossein Fatech, cofounder and CEO of DFT, and his staff.

Hossein Fatech

Company information is summarized here.

The data centers owned and operated by DFT are the following:

Location of DFT data centers: California, Illinois, Virginia, and New Jersey

Sizes of operating DFT data centers and their critical IT power support

As for the Santa Clara data center (SC1), it is huge. I often walk along one side of the data center to a sandwich shop, so I have a firsthand feel for its size. The next picture compares the USS Nimitz and SC1.

The comparison of DFT's SC1 and the USS Nimitz aircraft carrier

SC1 (1,099 feet) is slightly longer than Nimitz (1,092 feet), although the Nimitz is wider.

The layout of the SC1 is on DFT's website.  But it looks like this.

SC1 layout

SC1 consists of two symmetrical parts. In phase 1, only the left half is completed, for immediate occupancy, while the right half is semi-completed with a basic foundation (no raised floors yet.)

The design philosophy of the center is summarized here:

The design highlights of SC1 include: (1) stable IT load support via 18.2 MW ISO parallel UPS, which combines the system redundancy, isolation, and fault-limiting properties of isolated-redundant systems with the ability to spread system load evenly across all modules like paralleled systems, (2) reclaimed water for evaporative chillers, (3) an on-site substation with 3 30 MVA transformers, (4) 16 2.25 MW generators with selective catalytic reduction (SCR) that filters 90% of the nitrogen oxide from the exhaust, and (5) four points of entry for fiber from four different providers.

In addition, the data center is designed for LEED green and has a dashboard to monitor some of the relevant parameters, including temperature and power usage. The following slide shows temperature and humidity at data centers across a wide geographic area.

DFT expects PUE of less than 1.38, which is remarkably good in the leased data center environment. I always used to wonder how it looks like inside, and now I know.

In a recent Critical Facilities Roundtable Technology Group meeting, two speakers from Fujitsu gave interesting talks. The first gave Fujitsu’s take on cloud computing and was presented by Richard McCormack, SVP, Enterprise Systems, Fujitsu America. Most data center meetings are oriented towards facilities people, so a topic like this is rare. Richard was a good presenter and gave a lot of background on the thinking behind Fujitsu’s cloud computing strategies.

Richard McCormack

Before getting into the meat of the presentation, Richard shared some interesting data. An IDC list of the world’s largest IT services companies ranks Fujitsu third, with $24B. First is IBM with $56B, second is HP with $35B, fourth is Accenture with $22B, and fifth is CSC with $16B. Also, on IDC’s list of the world’s largest software companies, Fujitsu is ranked ninth.

Richard’s talk was organized to discuss changes happening in the IT industry, including changes in the business climate. He identified four major points of change and drilled down into each area lightly. The last point of change is the cloud, which was the main topic of the meeting in conjunction with their data centers.

Fujitsu’s understanding of how IT has progressed is as follows. It was first computer-centric (mainframes), then network-centric (distributed computing), and finally human-centric (information is all over the place but can be accessed anywhere). Changing IT are the six Cs:

Six Cs changing IT

Richard also said that the business landscape is changing, and he pointed out three changes: (1) the business border is disappearing, (2) a new generation of people is IT savvy, and (3) the consumer fields lead businesses in technologies.

Fujitsu’s overall vision of IT, consisting of people, information, applications, and the cloud, is illustrated in the following figure.

This view is self-explanatory. The cloud runs applications, which generate and/or collect data and convert it into information. That information is fed to people.

A slide explains each part. As for the people piece:

Again, you may have a slightly different take on this, but you do not disagree with the elements listed here.

As for the information piece:

Again, you probably agree with this description of the information piece. One interesting thing here is that Fujitsu develops and markets many products, including servers, storage, and networking gears. For identity management, Fujitsu has a palm reader that detects a unique vein pattern in your palm. Most palm readers read only the palm perimeter. Thus, if a hand is chopped off, as long as the shape of the palm is intact, you can use the chopped-off hand to gain access. However, vein pattern detection rejects the chopped-off hand. Fujitsu uses this palm reader for their data center access control.

More and more data and information are created from many sources. Fujitsu did a survey to find out how secure people feel about storing their data in the cloud. See here. The regional differences are interesting. People in the US care the most about where their data is, followed by Japan and Europe.

Fujitsu will publish another report on how much people care about sustainability at data centers from the IT perspective. The report states that interest in conserving energy is falling. The number one reason is that there is no easy and effective way to measure power consumption at data centers. It is a sad reality, but turning it around, if there is an easy but effective way to measure power consumption, people will conserve more.

The applications piece is further explained in the following picture.

Finally, the cloud piece is further described as follows.

Richard did not go into detail on each subject. Although you may arrange these items slightly differently from the way Fujitsu did, they must be relevant to you and your operations. For that reason, it makes sense to present a comprehensive view of the four visions: people, information, applications, and the cloud.

Richard completed the explanation of the "Delivering Change” figure with two slides on delivery of services and new technologies.

The rest of the discussion was about Fujitsu’s data centers. In short, they retrofitted their hosting data centers to:

• expand physical space

• increase power density

• increase high availability

• establish Tier III status

The actual change took place from August 2009 to March 2010. I will report on that in the future, along with their use of fuel cells for power.

This was a good presentation, well prepared and professionally delivered. The only complaint I had was the lack of discussion of Big Data. Fujitsu manufactures almost every kind of device and technology that contributes to the creation of a lot of data, such as mobile devices, Web and Internet devices (including servers and PCs), sensors and RFIDs, and automobile-related technologies and products. Also missing was information about analytics. If we receive a huge amount of data, how would we store and analyze it for our advantage? I would like to hear what Fujitsu has to say about that.

Since 2008, I have written several blogs to show the progress of this data center’s construction. When I started, I thought I would show the construction timeline when it was finally completed. Because of the stoppage of more than a year, I did not think construction would complete in 2009 or early 2010.

The following blogs touched on some of the construction timeline:

• DuPont Fabros has a financial problem (reported April 7, 2009)

• The suspension confirmed (reported on October 25, 2008)

• Weeds were everywhere at the construction site (reported on June 12, 2010)

• The construction resumed (reported on June 17, 2010)

• The resumed construction is at full-blown scale (reported on November 16, 2010)

• Near completion (reported on July 27, 2011)

I will report on the details of the data center later.

Recently, I had a chance to attend a Cloud SIG meeting held in the heart of Silicon Valley. The following information is not very technical but includes some history and an overview of what OpenStack is doing now. Specifically, I wanted to show who is behind OpenStack’s efforts. Although there are many more people involved besides the speakers, their pictures may bring it closer to you. The whole session was videotaped by the Cloud SIG of SVForum and may become available later, along with the presentation slides.

As reported elsewhere, Rackspace and NASA got together and developed a set of open source infrastructure as a service (IaaS) software packages.

Six people heavily involved in OpenStack gave short presentations:

As expected, people move around from company to company, and without knowing a little bit about each speaker's background, it is a bit confusing.

A Short Summary of OpenStack

Before getting into the meat of the discussion, let me summarize what constitutes OpenStack. There are three major components, but there are several more cooking. These are the three major pieces now:

• Compute (Nova), which manages things like provisioning to set up the computing environment.

• Object Storage (Swift), which provides something similar to AWS' S3.

• Image Service (Glance), which manages virtual machine images.

Other projects are listed here.

They are written in Python, and several people asked why Python instead of Ruby or something else. Joshua referred to a classified report submitted to the White House and defended his language choice but did not elaborate.

OpenStack is gaining popularity, and the community, including companies, is growing:

The list is growing, so you may want to check the OpenStack website for the latest.

Two Japanese companies, NTT Data and Midokura, were listed as code contributors. I was surprised, because it is rare that Japanese companies get recognition for their contributions to open source efforts.

Speakers

Rick Clark, now with Cisco, was one of the guys at Rackspace who conceived the idea of OpenStack and initiated the work with NASA.

Rick Clark

Vish Ishaya was with NASA Ames (hired by Joshua McKenty) but is now with Rackspace.

Vish Ishaya

Vish is in charge of the compute piece. He joined NASA to work on OpenStack and was with NASA at the time of the project start. He described what it was like at the beginning working with Rackspace. As he said, it is a bit strange now that he works for Rackspace.

Joe Arnold is with Cloud Scaling and is working on object storage for OpenStack.

Joe Arnold