In a recent blog post, Data Center Knowledge touched on AISO. The post included a video clip, produced by NetApp, in which Phil Nail, AISO’s founder and CTO, and Jeffery Byron, of the California Energy Commission, talk briefly. Since the video is in the public domain, I am posting it here for your convenience. It runs a little more than three minutes.

What AISO does is simple. They use solar power generated at their data center, keep rainwater, and use it for cooling. In addition, they reduce IT power consumption by applying deduplication and other consolidation technologies from NetApp, IBM, and VMware.

The California Energy Commission states that 2011 is the year to begin serious energy efficiency efforts at data centers. Such measures will be implemented in 2013, saving as much as $800M in energy costs.

What AISO does is commendable. However, we cannot expect their practice to be easily duplicated elsewhere. First, not every site has the luxury of enough solar or wind power. Even if it did, the power generated might not be enough to run your data center, depending on what you run. Second, if you don’t have a good energy storage solution, the intermittency of renewable energy sources could keep you from sustaining your operations 24/7.

As I look at microgrids and distributed generation, I don’t see many that rely solely on renewable energies like solar and wind, except in some experimental cases. There are far more microgrids and distributed generation fueled by natural gas. With the natural gas turbine, you can control the amount of power generation with relative ease, and you do not have to worry about the intermittency of wind or solar.

We sometimes have the notion that microgrids equal renewable and clean energies. But except in some specific cases, natural gas is the dominant source of energy for microgrid. I will cover microgrid more extensively in future posts.

Those data must be aggregated, processed, analyzed, stored, and queried. Some data may be shared by multiple applications. Some applications interact with other applications via data and/or control. Even before discussing specific applications, we know the following:

• Multiple applications must be loosely integrated to exchange data and control, and that requires an enterprise bus.
• Each application and each platform for applications must be expandable and scalable. We cannot foresee all the future applications that might require specific interfaces and high scalability. It is very hard to predict what they will be, but we can make the architecture as flexible as possible.

Data can be roughly classified into two types: real time and non–real time. Monitoring of transmission and distribution lines happens in real time, and the collected data should be processed in real time so that any necessary measures can be taken. In severe weather conditions, outages could occur at any second, so the data associated with such weather also falls into this category. However, other weather-related data, such as tomorrow’s temperature, may not be processed in real time, although prompt processing may be necessary.

Another type of real-time information concerns generator status. Several different types of generators are in service at a given time. The attributes of each generator are in the utility’s database. Such attributes include total generation capacity, operation requirements (the number of hours the generator is permitted to run before it hits the limit on GHG emissions), current health, and any contractual agreements of its owners (if the generator belongs to someone else). In high demand time, the utility needs to assess what to do. It may calculate the number of kilowatts to curb by issuing D/R signals and seeing if that would satisfy the current demand. If not, it may have to start its own or someone else’s generators to satisfy the demand.

The total power consumption data also must be processed in real time to balance the fragile demand/supply balance. But the same data are also processed in non–real time for consumers to obtain. One of the biggest advantages of smart grid is that it informs consumers of their hourly (or less) power usage. Each household’s power consumption data are collected along with millions of other households’ and sent over to the utility’s back office to be processed and stored. Each consumer has access to the stored data. Currently, in the PG&E territory, access to the data is at least 24 hours late, and power usage information is only hourly. Some people claim that more frequent usage data is desirable. However, I am not sure what to do with data collected more frequently than hourly.

Several applications and keywords stand out in smart grid reports, articles, and books: SCADA, DMS, EMS, OMS, MDM, DA, grid optimization, AMI, CIM, D/R, and so on. I am not going to touch on each one here. But all these applications still deal with fairly low layers of smart grid and are directly related to the health of the infrastructure and the data collected by the meters. As smart grid progresses, more applications not directly related to usage data or infrastructure health will be developed. Everyone wants to know what the killer applications are going to be. At this time, D/R is considered a killer application. D/R is like magic. It produces extra power without your having to turn on extra power plants or generators. Yet it fixes the fragile balance of demand and supply without spending a lot of money, and does so almost instantly.

If you can name the next killer applications, you could build a Google of smart grid and become very successful.

• The distribution grid is much longer than the transmission grid. The total length of the transmission lines in the U.S. is about 300,000 kilometers (roughly 186,000 miles), while the distribution grid runs several thousands of miles.
• The transmission network carries power at high voltage (xx V). The distribution grid carries a much lower level of voltage ( xx V).
• In general, the transmission grid is more reliable, having rigorous monitoring and control by means of sophisticated computer systems. However, little monitoring is done for the distribution grid, which is aging rapidly without replacements.

Even though the two networks are very different, each requires managing and controlling functions like supervisory control and data acquisition (SCADA).The two SCADA systems are not identical. ABB’s website has a good description of each.

SCADA/EMS (Supervisory Control and Data Acquisition/Energy Management System) supervises, controls, optimizes and manages generation and transmission systems.

SCADA/DMS (Distribution Management System) performs the same functions for power distribution networks.

Both systems were developed with many modern ICT technologies. Measured and collected data must be transmitted in real time. The improvement in sensor technologies, semiconductors, embedded systems, communications technologies, sophisticated software, and more has contributed to the availability of these systems. Also, IP is becoming the network protocol of choice over proprietary ones. Of course, an extensive power industry and power engineering are required to develop these systems. System integrators with domain knowledge in power, like, perhaps, Perot Systems (recently acquired by Dell), are good candidates to develop and maintain SCADA systems, in addition to ABB, which is an expert in power and control systems.

SCADA use in transmission lines is very common, but there are at least half a dozen versions used in the transmission network, according to Peter Fox-Penner.

This prevents the players in the transmission network from sharing the status of the networks in real time. Moreover, another big improvement in this field is the addition of synchrophasor technology (also known as phasor measurement units). This technology may belong to power engineering but not to IT. However, with the improvement in the technologies I mentioned above (regarding the advancement of SCADA), this technology is now better and available.

On the other hand, the use of SCADA in distribution has just begun. In recent blog posts and articles, some analysts are starting to say that utilities should work on improving the distribution network because the end-to-end network is in place (or being put in place), connecting generation to consumption.

This article mentions Current Group, which provides real-time sensing and control technologies, which fall into IT technologies.

I accept that the transmission and distribution field is oriented more toward power engineering than toward ICT. However, it is also true that the improvement in many ICT technologies has helped advance the transmission and distribution technologies.

Smart grid introduced the term FAN. FAN is part of AMI, the infrastructure connecting the end consumer to the utility. A smart meter aggregates home power consumption data and passes them on to a FAN, which then transmits that data to a WAN, which delivers them to the utility.

For the average consumer, the smart meter is the gateway to the utility and the interface with the home area network (HAN). Although there are other gateways, such as broadband cable, DSL, and fiber modems and connectors, the smart meter is gaining a foothold as a gateway. Since the utility installs it, the smart meter is already compatible with the utility’s communications protocols. For that reason, it is becoming the de facto gateway.

The meter technology is unique, and I am not sure whether typical ICT vendors can enter the field easily. Some vendors, like GE, stand out among the who’s who in providing meters. In my PG&E territory, we had an analog (dumb) meter from GE before, and our new smart meter is also from GE. A big difference between dumb and smart meters, aside from analog vs. digital, is the communications capability of the smart meter. The smart meter has at least three functions: it aggregates/stores the power usage information (hourly or more frequently), transmits the aggregated data to the utility via the FAN, and receives the signal from the utility to control devices and appliances on the HAN.

A dumb meter stores the power usage information by advancing hands in the analog indicators, and it needs no specific internal memory. On the other hand, a smart meter requires some kind of internal memory to record the usage according to the frequency of measurement. In addition, a smart meter needs a mechanism to transmit and receive data and signals. A transmission chip or firmware is embedded into the meter logic for that. ZigBee (mesh wireless technology) is becoming the de facto standard for smart meters. (ZigBee is not a company but a consortium that dictates the ZigBee specification. A company that wants to implement ZigBee pays a license fee and implements its own version according to the specification. NIST is very cautious about declaring a technology a standard. For example, NIST has designated IP as a communications protocol but no others yet. ZigBee, Wi-Fi, and WiMax are under consideration to become part of the standard. ZigBee has started working on supporting IP but doesn’t do so completely yet.

Data collected from each household are aggregated to a network access point (NAP) in the neighborhood. I have been looking for this point but have not found it yet. Since ZigBee does not support IP 100%, the communication between each smart meter and the NAP is via ZigBee.

Some information about PG&E’s smart meter can be found here.

Each NAP is now connected via wireless WAN. PG&E uses ZigBee for FAN, but other technologies could be wired  or wireless (WiMax) and BPL.

The power system consists of generation, transmission, and distribution. Let’s start with distribution, which also has three parts: the distribution network, field area network/automated meter infrastructure (FAN/AMI), and the home. Today I’ll talk about the home aspect.

A smart meter installed at home provides power usage information as an aggregate. By observing my hourly power usage information, I can understand my usage. (By the way, I do not think we have any dynamic pricing yet, as the usage and the charge for it are proportional.) This is great progress, but I would like to see a more detailed breakdown. For example, there is a spike in the middle of the night when I shut down all but essential appliances like the refrigerator. I would like to know if this spike comes from the refrigerator’s defrosting cycle or something akin to it.

To support more-detailed information, each appliance and other electronic device must be able to report its power use to the smart meter or some other collection point in the house. To do this, we need a module to meter power usage as well as some kind of communication function. A new set of appliances and electronic gear may come with the metering chip or firmware installed, but we need a dongle for the existing ones. The dongle business would be only until our current appliances are all replaced by new ones that have the chip or firmware. As for communication, we need a mechanism to transmit the power usage information to the smart meter and/or some collection point. The communication should be two-way because, when demand and response (D/R) is implemented, each appliance and other electronic gear must receive a D/R signal and respond to it. Whirlpool, for example, has announced that it would include such a function in all its new appliances. The candidates for the communications technology include HomePlug, ZigBee, Wi-Fi, Z-Wave, and 6lowpan. None of them is designated as a standard by NIST, though.

Right now, the only way I can get my power usage information is to access PG&E’s website. The information is delayed more than 24 hours. Even though this is progress over the one-month-late information on the bill, I want to receive more minute and real-time information. Unless the data sent over to PG&E are available to me, I cannot do anything with them. However, if its own collection module is embedded in each appliance and electronic device, any power usage information, whether aggregate or individual, can be readily available. The metering frequency can be adjusted at our discretion. Once such data become available, they can be displayed on any of several dashboardlike software applications now available. Google and Microsoft provide free software, while companies like Opower sell their own versions. Opower works with utilities to provide its product. PG&E has not worked with either Google or Microsoft, stating that it wants to wait for the standard.

The home market may be large, but the technical barrier to entering it does not seem to be too daunting. The chip or firmware for metering power usage is a commodity. The communications protocols are well known and do not appear to be hard to implement. Once the power usage data are available, collection, aggregation, analysis, and display are very straightforward, and I do not see much differentiation in the technology itself. So probably the key to this market is how well each vendor and service provider can work with utilities. After all most consumers are not engineers and would like to have an easy solution. On top of that, power consumers deal with their utility, and adopting the solutions provided by their utility may feel easier.

By the way, as one of the speakers at SVLG’s energy summit, I find my interest in checking the hourly use of power waning. After all, I cannot see the breakdown or real-time usage. I will talk about the FAN/AMI area tomorrow.

Looking at the Data Center World program, I can tell what’s hot in the data center industry: (1) cloud and data centers, (2) ITIL, (3) the colo decision (to buy or build), and (4)cybersecurity. ITIL stands for Information Technology Infrastructure Library and is a set of concepts and practices for information technology services management (ITSM). ITIL is being considered to formally define processes and activities for data center operations.

At the last Data Center World conference in Nashville last March, this subject gained some attention. This time, there is a tutorial on it. The interest in cloud computing and how it affects data centers remains strong. Economies of scale really work—unless you build a really large data center, it does not make sense to run your own. Two things encourage colos: (1) data center construction is capital intensive and (2)time to market is long if you build your own. Security is becoming a big problem as more data and connections are brought into data centers and cloud computing enters the mainstream, which is a mixed environment.

The third annual SVLG Data Center Energy Efficiency Summit will be held at Brocade’s new headquarters in San Jose. As usual, real case studies of energy efficiency at data centers will be presented. AltaTerra will publish a summary report of this summit as we did for the sustainability conference in March and the energy conference (to be published soon).

Facebook users increased by 100 million in the first seven months of this year (January to August 2010). I can see their need for more data center floor space. Greenpeace is not happy about the location Facebook selected, because the electricity source is coal in that region. Facebook shot back that they are using waterside economizer (evaporative cooling), exploiting the region’s dry and moderate climate.

Selecting a data center site involves considering several factors, including power and fiber. There are many other points to consider, such as taxes, staffing, proximity to clients, land cost, and clean energy sources. I suppose Facebook made its selection after considering those factors. It is easy to criticize others, but when you do, you need to clean your nose first. Remember that Greenpeace’s data center itself was not exactly green?

I am sure many data centers operating on a smaller scale are much less energy efficient. Remember, some 90% of U.S. data centers are either server closets or server rooms. I doubt that they pay any attention to energy efficiency at this time. Greenpeace targets Facebook because of its high visibility. But there are many more data centers that can be made more energy efficient.

Now that Oracle has Sun’s hardware, Solaris operating system, virtualization engine, and other components, they can provide well-integrated solutions for data centers, as this slide shows:

The plan is to increase server and storage performance by several times and even tens of times more in a few years.

Server Performance Increase

Storage Performance Increase

In addition to Solaris, they have Red Hat–compatible Oracle Enterprise Linux (OEL) and support. Their virtualization is implemented through Oracle’s version of Xen software, Oracle VM. On top of Xen, Oracle added management and other software to make it a comprehensive offering for virtualization. The features of Virtual Iron, which Oracle acquired some time ago, are being integrated into Oracle VM. Both Oracle VM and Virtual Iron are based on Xen, so the integration should not be too hard.

A presentation on cloud computing was a good tutorial. Many of Oracle’s customers are enterprises, which tend to use VMware’s virtualization solution. Oracle provides a feature to translate VMware VM file formats to theirs. Currently, private or on-premise clouds are implemented mostly with VMware, and public clouds (AWS and Rackspace) are implemented with Xen. Eucalyptus’s enterprise version has a feature that is the reverse of what Oracle VM does. It can translate VMs by VMware on-premise to AWS file formats to allow them to be transported to AWS public cloud (known as CloudBurst).

Oracle view on Enterprise Evolution to Cloud

Since Oracle now owns microprocessors, server and storage hardware, operating systems, virtualization engines, databases, and applications, they can fine-tune the entire system to make it very efficient and execute very fast. Although Oracle is still small compared with IBM in terms of revenue (Oracle $27B vs. IBM $96B), Oracle is now in a position to compete with IBM.

The seminar was well run and presented very useful information. If there was one thing I did not like about it, it was the lack of discussion of energy efficiency. Energy efficiency and green IT initiatives were mentioned several times during the day, but they were not discussed in detail. I asked one of the speakers to share some energy efficiency data with me. If and when I get it, I will publish it here.

Picture of the construction site

Time passed, and every time I drove by I saw more weeds growing and covering the half-completed structure. Suddenly, I see some movement again. The construction is going rapidly. I see a lot of people on the site working with heavy machinery.

As has been said, the data center space available is not at all adequate to the demand. Much construction was put on hold in 2008, and now that the economy is picking up, there is not enough space. It takes 18 to 24 months to build a new data center from scratch. I understand both Google and Microsoft are completing and extending their data centers.

Because of time to market and the cost of construction (hundreds of millions of dollars), enterprises are giving more consideration to colocation and wholesale data center arrangements. This trend certainly helped Dupont Fabros, and I will keep watching their progress here in Santa Clara.

If the AC power entering the data center is converted only once to DC power and distributed to IT equipment that takes DC as input, there would be no conversion loss. Unfortunately, a few years ago, Green Grid put out a white paper comparing the power distribution of DC and AC and concluding that there were very few differences between the two configurations. After reading it, I moved my focus away from power distribution. Now I am back on this subject because some friends who started a company in that area and Keizo Hoshijima of NTT Facilities pitched me the merits of DC distribution recently.

I cannot forget another DC power distribution advocate, Dennis Symanski of EPRI. I met Dennis when he was a panelist in my Nordic Green panel session.

Dennis Symanski

At the conference, he gave a presentation on DC power distribution. Cooling is known to consume about 30–60% of power in a data center, but we seem to be getting a handle on that. Once cooling and other culprits are under control, then we can pay attention to power distribution to further cut power consumption.

Recently, I had a chance to visit Dennis at EPRI to chat about the current status of DC power distribution at data centers. The following is an edited summary of our conversation.

Q: What do you do at EPRI, and what is your background?

A: My focus is to make any computer equipment energy efficient (EE). Prior to coming to EPRI, I spent 18 years at Sun working on international standards and regulations. EPRI is funded by utilities but is an independent nonprofit research organization. It strives to "do good for society,” and it is quite refreshing for me when I do not have to pay attention to revenues and stock prices.

Q: So what’s new in the DC power distribution area?

A: In 2006, we conducted a set of experiments on the use of DC power distribution at Sun (currently Oracle), along with LBNL, Ecos Consulting, and other vendors. Our conclusion was that DC power distribution increases data center EE and reliability. Since the experimentation, we formed an organization called DC Power Partners to further our efforts. The members include LBNL, EPRI, EMC, IBM, HP, Oracle (former Sun side), Intel, and even APC. (ZK; APC had a different opinion of DC power distribution before.) We have a once-a-month teleconference to discuss new technologies and installations to share information. In addition to the U.S. companies, some European and Japanese companies call in.

Q: You only cover technology aspects at your DC Power Partners?

A: EMerge Alliance works on DC lighting, led by Armstrong Ceiling for 24 VDC. DC Power Partners is joining the alliance. They can take care of sales and marketing, trade shows, and web promotion.

Q: Are there any major changes in the DC power distribution area?

A: There are no drastic changes. We are currently working to make sure the components that comprise DC power distribution comply with UL and/or FCC regulations. For example, connectors that are used for DC power distribution.

Q: I have no issue with the technology. What about the market? Which region of the world is more receptive to this technology?

A: In general, the telecom industry is keen on picking this up. Companies like AT&T, Verizon, and NTT. In Europe, the European Telecommunications Standards Institute (ETSI) has prepared a specification for 380 VDC. There is at least one CO in each city and dozens in large cities, so there are 10's of thousands of COs in the U.S.

Q: But IT-centric data centers and COs are different, and the IT data centers may not be so enthusiastic about DC power distribution.

A: They used to be quite different, but, these days, they are becoming increasingly similar as they share the same kind of IT equipment. Look at AT&T, which needs to process IP traffic coming from iPhones and iPads.

Q: A couple of years ago, I read a white paper from Green Grid (GG) that said there were few differences between AC and DC power distribution. What is your take on that?

A: GG members do not include any direct current facilities equipment suppliers. Several of the GG IT equipment suppliers are researching direct current power supplies for their equipment.

Q: What do server vendors need to do to support DC power distribution?

A: All they have to do is to swap the AC power supply with the DC one. The direct current power supplies have a smaller component count, are more efficient and as a result, research demos may show them to be more reliable. It is not hard to create a data center solely with DC power distribution. IBM created such a data center at Syracuse University.

Q: Is there any data center with a solely DC power distribution system in the San Francisco Bay area?

A: Several are going in parallel. But one in the Bay Area will be announced in November.

Q: Right now, generated power is carried via the AC transmission system. If DC power enters a data center, we do not need to convert power at all. Can you transmit power via DC?

A: In the days of Edison vs. Westinghouse, we could not transmit DC power at 5,000 V or higher. These days the high DC voltage is transmitted over a long distance. A hydropower plant in northern Quebec transmits generated DC power to New England over several hundred miles, in addition to much-closer Toronto and Montreal. (ZK: High-voltage direct current is a technology for transmitting high voltage DC over thousands of miles)

Q: Is there anything else I should know about what EPRI is doing in the DC area?

A: We talked about the power supply for IT equipment. Those power supplies were very inefficient. The conversion rate was only 65% on the average. 80 Plus is an organization that promotes increasing the conversion rate to at least 80%. EPA adopted their specification, and EPRI is running a test lab for 80 Plus.

I got very useful information talking with Dennis. EPRI covers a large area of energy and its efficiency. I have not heard much about DC power distribution recently, but that does not mean efforts on its behalf were terminated. On the contrary, they are going very strong, including DC for home and power transmission. I will report on this subject from time to time in this blog.

In a slide, Zahl Limbuwala of BCS’s Data Centre Specialist Group presented a few interesting metrics, including the Reuse Energy Ratio (RER). Since I was not familiar with RER, I asked Zahl what it is all about. It turned out that this metric is being kicked around only internally within The Green Grid, and no public information is available. The BCS Data Centre Specialist Group is working with The Green Grid on the metric but is not at liberty to discuss the details of it.

They keep it pretty private because the only thing I can find out about RER is from a presentation by Dan Azevedo, chairman of the Metrics and Measurements Work Group at Symantec. 

Even this presentation mentions RER only once, without details.

By the way, Zahl gave me two of his research reports on metrics and will give me a new one when it comes out soon.

PUE considers the lump sum of power usage by the entire data center vs. the power used by the IT equipment. It is not hard to think that the data on IT equipment efficiency and utilization, along with the outputs, should be considered in profiling the energy efficiency of a given data center, which consists of facilities and IT equipment. More to come in future blogs.

Marten Mickos

The following is a summary of our talk.

Q: Correct me if I’m wrong: Eucalyptus provides a set of open source tools and utilities to create on-premise clouds. All the components are either from Eucalyptus or some other open source. In addition, it has a hook to pick different types of hypervisors and to run on major Linux distributions like Ubuntu and CentOS. It is largely compatible with the Amazon Web Services (AWS) environment. You can move your virtual machines (VMs) between Eucalyptus and AWS clouds.

A: Basically, you are right. We provide a platform for building on-premise cloud.

Our enterprise version supports multiple hypervisors, such as Xen, KVM, and VMware.

Q: Can you name your customers for the enterprise product? Is the pricing based on number of CPUs?

A: We have several customers that are large enterprises and government agencies. I cannot reveal their identities yet, because I need their consent. Licensing is based on cores and we charge about $300 per core.

Q: What is an on-premise cloud, and what is the difference between on-premise and private clouds? In the past, you were looking for a better term than on-premise cloud. Did you find one?

A: An on-premise cloud runs on your hardware infrastructure at your site. A private cloud is a cloud reserved for the use of one single organization. In many cases, if not all, you can use the two terms interchangeably.

Q: What is your license for the open source side?

A: It is GPL version 3.

Q: Who develops code and commits it to the main trunk? Who has the copyright of the code? Do you take others’ contributions as well?

A: We develop our code and have the copyright of the code but take in others’ contributions as well.

Q: So this is somewhat similar to what you did at MySQL?

A: Yes. But the difference is that MySQL was disruptive in the old database market, but Eucalyptus is innovative in the new market of cloud computing.

Q: You are a big proponent of open source, and you like what Rackspace recently did? I mean that it created OpenStack for clouds. NASA’s Nebula project uses Eucalyptus, and they contributed their code to the OpenStack project. Did they include Eucalyptus in OpenStack as well?

A: Yes, I like what Rackspace did. I do not think they included Eucalyptus, because our license is GPL and OpenStack is by Apache 2.0. With open source, users can naturally take your code, use it, modify it and redistribute it. In the cloud market there are new projects and companies being launched all the time.

Q: If everyone can do it, what you are doing? What are your differentiation and/or advantage points?

A: We see it as a benefit, not a problem, that open source gives users many freedoms. Already today we have tens of thousands of users who do not need our commercial support. But the most mission-critical installations are dependent on our support, bug fixes and further development of the product. Uniquely in this market at this time, we have a scalable, mature and functional product.

Q: Amazon uses Xen open source, yet has its own APIs and file format for VM. So open source is closed by a proprietary container and is no longer open source. What do you think of it?

A: Again I think it demonstrates the power of open source. Open source is needed in all major infrastructure deployments today. What Amazon is doing is perfectly within the licensing terms and principles of open source.

Q: What is Dr. Rich Wolski (the founder of Eucalyptus) like? Is he a typical researcher?

A: Not at all. The typical researcher does not produce very useful things like Eucalyptus. Also the typical researcher does not start a new business.

Q: Is cloud computing more energy efficient?

A: Yes it is more energy efficient. But we must remember that certain computation always requires a certain amount of energy. It is the fact that you have higher utilization rates in cloud computing and you can turn off servers that are idle for some time to save energy.

Q: What about the opposite of that? What if more demands come to you beyond your capacity? Do you load-balance among several different clouds? If so, what are the problems in doing that?

A: Yes, the vision of cloud computing includes cloud bursting, which allows you to move load dynamically from one cloud to another. Technically it is an advanced proposition and we are not there yet for the general public. There are intricate challenges around security, authentication and latency that the industry is working on.

Q: What geographical markets are you addressing now?

A: Four regions are very important: the U.S., the E.U., China, and Japan.

Q: Are you doing anything in those areas, other than the U.S.?

A: Eventually, we will be in all of the areas, but right now we have our hands full with the U.S.

Q: We are sitting and talking here in Silicon Valley. Do you plan to move Eucalyptus to this area?

A: In the days of cloud computing, the physical location of headquarters does not make a difference. The headquarters is where the CEO is.

Q: So it is like Air Force One. When the president gets on the plane, that plane becomes Air Force One.

A: Exactly. Now this cafe is Eucalyptus One, and I can make all the executive decisions from here! In reality, of course, a company makes decisions in many locations and by many executives. At Eucalyptus we work as a team and for this analogy to be perfect, we need to consider the locations of every employee of the company.

Q: One thing I remember about you, during your days at MySQL, is that you came up with good analogies. You said that regardless of which seating area (first class, business class, or economy class) in an airplane you are sitting in, everyone gets to the destination at the same time. This was in reference to MySQL vs. Oracle databases. Do you have a good one for cloud computing and/or Eucalyptus yet? That kind of description grabs people’s attention quickly and easily.

A: I am thinking about it but have not come up with anything yet.

I had not seen him for a while, but Marten remains as he was before, very nice and friendly. As I researched into Eucalyptus, I found many similarities with MySQL. They are both based on open source, but the business model has a commercial version as well. MySQL was run as a virtual company, which, come to think of it, is a cloud in a sense.

Cloud people tend to ignore the underlying infrastructure; data centers. Marten has a good sense of what is below the cloud. When he talked about turning off servers in the case of low loads, I thought about Power Assure, which has a feature to turn servers on and off as needed. If you can synchronize the operation of IT and facility (cooling and power delivery) equipment according to the loads you receive, you could save more energy at your data center.

I asked Dennis for his opinion. He told me that he recently gave a talk internally at EPRI regarding the types of data centers in the U.S. He gave me the following handout, which shows the sizes of data centers in the U.S.

IDC created this table, which classifies U.S. data centers into five categories and estimates the total number of data centers and servers in each category. This data was as of 2006, and IDC predicted what it was going to be three years later. This table does not reveal much interesting information at first glance. So I have created two graphs: the first for the number of data centers and the second for the number of servers in each data center.

The first graph shows the relative ratio of each data center type. The vast majority of existing data centers are either server closets or server rooms.

The number of servers for each data center is shown below.

The servers are almost equally divided among the five categories. This means the larger types install a large number of servers.

So Dennis’s theory is that the vast majority of data centers are small and have no one specifically in charge. If that is the case, it would be very hard to exercise energy efficiency technologies or practices.

I will report in the near future on our conversation on how to use direct current power distribution at data centers.

At the recent DatacenterDynamics conference in San Francisco, Zahl Limbuwala, chairman of the Data Centre Specialist Group at the British Computer Society, moderated a panel session titled "Industry Executive Roundtable—Are we Reaching an Inflection Point in Our Industry?” He talked about the U.K.’s Carbon Reduction Commitment last year at the same San Francisco conference.

Zahl Limbuwala

I will report on this roundtable session in the future, but for now I want to bring a very interesting slide to your attention. The usual presentation has a slide showing a list of metrics. But Limbuwala’s slide goes farther to depict what each metric addresses. See the slide here.

One thing missing is CADE, which I talked about along with the Japanese-proposed DPPE.

I was not very familiar with RER, and when I searched for the term, it got only two hits. I need to talk to Limbuwala to get details. For that matter, I will ask him to see if he has a paper or two on the metrics. In any event, this slide is very good at giving an intuitive view of which factors are considered by which metrics. According to this slide, DPPE takes the most factors into consideration. Whether that is good or bad is still debatable.

The panel consisted of:

Moderator

• Dan Geiger, U.S. Green Building Council

Panelists

• Steve Berkman, Paul, Hastings, Janofsky & Walker
• Meade Boutwell, CB Richard Ellis
• Mark Bramfitt, Bramfitt Consulting

From left: Dan Geiger, Mark Bramfitt, Meade Boutwell, and Steve Berkman

The U.S. Green Building Council is an organization that sets LEED, which is a green-building certification system. The current version of LEED is the third, set in April 2009. Although a data center is categorized as a building, it is very different from ordinary buildings. Because of that, some efforts have been made to establish a version of LEED applicable to data centers. There are usually two to three years between versions of LEED, and the next version, covering data center, retail, and hospital considerations, will not be published before next year. Geiger spoke of the growth in the green building market. It is $71B this year and expected to grow to $173B by 2015, including new construction and retrofits. The council is working with LEED-certified building owners on metering and measuring building data. Initially, they want to collect energy and water data and analyze them for even better efficiency. Later they would like to include waste data as well.

According to Boutwell, the California real estate market, like the rest of the country’s, is in bad shape, but in the long run the market will come back. In five years or so, most buildings will require some kind of green building certification like LEED. Incidentally, the housing market recently began growing again in my neighborhood (the South Bay, in the heart of Silicon Valley). The selling prices are still a little lower than those at the peak, but they are coming back steadily.

I always like Bramfitt’s presentations, and he did not betray my expectations this time either. Prior to this panel, Simon Husha’s talk on the scaling of data centers indicated that California is not considered a desirable place to have data centers. Bramfitt disagreed with that assessment, saying that for the past five years data centers have grown by 15% in the state, especially in the city of Santa Clara, and mostly by colocation.

Asked about Title 24, the California Energy Commission’s standard for energy-efficient residential and nonresidential buildings, Bramfitt said that it does not include data centers. So he is working with utilities and other groups to include data centers in Title 24. Although the final version is not set or approved, he gave us a hint of what is to come. Since the original standard encourages energy efficiency for residential and nonresidential buildings, the data center section might include something like the exploitation of environmental conditions for cooling, such as by air or water economizer, and more energy-efficient power distribution systems and UPS.

He said it typically costs a data center operator an additional $800K to build the airside economizer function. PG&E may give a rebate of $400K, and the operator enjoys a saving of $1M a year. To Bramfitt’s amazement, some data center operators still do not go for that option.

Regarding how to control climate change, he thinks cap-and-trade will not succeed and something like a carbon tax will be imposed and passed on to consumers. Someone suggested that data centers could be built in a moderate climate to exploit lower temperatures for cooling. Bramfitt said that in his experience, the #1 criterion for selecting location is power cost. If the power cost is very low, the operator usually does not pay any attention to energy efficiency. For this reason, Bramfitt thinks incentives alone will not guarantee energy efficiency. He thinks regulatory pressure works better.

He also mentioned that California has been a trendsetter, and sixteen other states usually adopt the standards California sets. So once this is set, it may spread as a nationwide practice. He also emphasized that the Title 24 that will include a section for data centers will be for new construction only and not for existing data centers.

Berkman pitched a few things into the conversation. As an attorney, he thinks it is a good thing to make buildings and data centers green. But it costs someone to implement green buildings. There are two different situations. One is that you have an agreement with your landlord before the green law is put in place. The other is that you need an agreement with your future landlord after the law passes. Who is responsible for the additional cost to green the buildings for each case? You need to pay attention.

From the market perspective, there is enough capital available on the supply side, but rather than building new data centers, companies are making more acquisitions. The ROI appeals to investors in the data center market. As for the demand side, during the past five years, Berkman has observed strong growth. And prices went up by 30–40%, with some sophistication, along with upgrades in services (like from managed services to colos, and colos to wholesale solutions).

Now I realize why Bramfitt’s session is always interesting. He provides very interesting data from the utility perspective that are usually not well known. On top of that, his delivery and the way he talks attracts the attention of audience. Cap-and-trade and policies similar to that have been put in place in Europe, the U.K. (most comprehensive), and Japan. I don’t know their status, but I recently heard several people talking about some alternative policies that might work better in the U.S., such as carbon taxes. It is still too soon to see how things will turn out. Even though the current version of cap-and-trade died, as long as the Supreme Court’s ruling holds, some kind of energy efficiency control will be imposed in the not-too-distant future in the U.S., according to Senator Maria Cantwell (D-Washington), speaking at the recent SVLG Energy Summit at Stanford University.

I visited Broadcom to meet with David Berry, senior marketing manager, and Wael William Diab, technical director, office of the CTO. David and I have talked about the general energy efficiency of networking equipment before. Wael is vice chair of the IEEE 802.3 Working Group

So I focused on finding out what 802.3az is all about. My core technology area started with software engineering and expanded to cover networking and embedded systems. I am by no means an expert in networking technologies but am dangerous enough to know some of the basics. I here confess that while I was summarizing my conversations with these two gentlemen, I studied and researched this subject extensively so I could keep up. If the whole thing is completely over my head, it is easy to let it go, but with some effort, I can understand the subject matter to some extent. Sigh.　　

Wael Diab and David Berry

The following is an edited version of my questions and their answers.

Q: My understanding is that energy efficient Ethernet (EEE) allocates networking resources when needed and turns them off when not needed. Can you give me a little bit more on EEE?

A: Traditionally, the most important things about networking were speed and cost. But energy consumption is becoming more important than either of those because opex is beginning to surpass capex. The Ethernet has typically been operated as always-on, even though traffic is not always on. A typical utilization is less than 10% and sometimes even less than 1%. But when traffic is back on, it tends to hit the peak, and the Ethernet has to quickly come back on at 100% capacity. When the PHYsical layer (layer 1 on the OSI model) is on, the layers above (layer 2, data link, and up) are on as well. When there is no traffic, the PHYsical layer (PHY) can be turned off, along with the layers on top of it, creating further energy efficiency. Additionally, we have implemented a way to negotiate when to turn on layer 2. In this way, layer 2 can sleep until absolutely necessary.

Q: Is the ratification of 802.3az done? Is the product based on the standard shipping?

A: The standard is at the final stage, and ratification is expected in September. However, prestandard versions are shipping by vendors, as has happened for other networking products before.

Q: Can this technology be implemented with firmware refresh?

A: Unfortunately, to support firmware refresh, a new PHY is required and, therefore, new equipment. We have a technology called AutoGrEEEn. AutoGrEEEn technology enables a device with a non-EEE MAC to seamlessly transition to EEE capability by implementing control policy assist engines and circuitry inside the PHY device.

EEE requires control for the PHY to be done via in-band signaling over the MAC/PHY interface. This requires a change to both the PHY and MAC silicon. A number of systems have the MAC and PHY as two different pieces of silicon with the MAC often embedded in a switching or controller-type device. These MAC-containing devices have associated drivers and software, and are often multiport devices. So, a transition to EEE may be hampered by additional development that involves replacement of the MAC-containing devices. AutoGrEEEn technology eliminates the need to change the MAC/PHY interface on the MAC silicon, and allows for rapid transition today with legacy non-EEE MAC silicon attached to AutoGrEEEn-enabled PHYs.

Q: Servers are refreshed every three to five years at data centers. How about networking equipment?

A: In general, the networking equipment refresh cycle is probably longer than that of servers. However, it is important to consider networking in conjunction with servers and other IT equipment. Servers get more powerful, with new chips every 18 to 24 months. For example, a newer server with more computing power may require 48 ports rather than 24 ports in the networking equipment to process more loads. Thus, the refresh cycle may not be that much longer than that of servers. In addition, new builds are happening all the time and they can take advantage of any new innovations in energy savings.

Q: I think 802.3az is for 10 Mbps, 100 Mbps, 1 Gbps, and 10 Gbps. Does it apply to 40G and 100G as well?

A: For 40G and 100G, there is another standard called 802.3ba, which was recently ratified.
Both 802.3az (EEE) and 802.3ba (HSE) started about the same time. EEE looked at copper
interfaces that already existed ton be enhanced for energy efficiency. By the way, the
energy efficiency technology we use is called
low power idle . When it is idle, it sends little energy and is very energy efficient.

To achieve this both sides go into a suspended state, effectively turning the PHYs off. To keep both sides synced at the PHYsical layer a refresh signal is occasionally sent. Because the refresh signal has a low duty cycle, it is very energy efficient. To leave the LPI state, either side can send a wake-up signal and the other side reacts to that.

In addition to PHYsical layer savings additional subsystems above the PHY can be turned off. Broadcom championed the development of a technology called layer 2 data link LLDP stateful negotiation for enhanced saving modes in the standard.

Q: Some people insist that saving on the networking side does not make a huge impact on data center power consumption. Some studies indicate that power consumed by networking equipment at a data center is only several percents compared with those of servers and storage equipment. What do you say to that?

A: It is true that the networking equipment by itself does not consume a lot of power from the perspective of a few single ports. However, this energy efficient technology (EEE) for networking enables energy efficiency across the entire data center which typically has 1000's or more ports. Enabling EEE is like turning off a bunch of leaky faucets. One might not notice the savings at a single point in time but it will certainly show up in the monthly bill. Ethernet has also been proven to be an extremely scalable technology enabling port consolidation during bandwidth migrations cutting total power consumption (1G to 10G for example) while at that same time increasing performance typically by a factor of 10. In a sense, Ethernet networking is playing a much larger role in turning a data center into the next-generation energy efficient one.

Think of this analogy. When you try to establish a high quality of service (QOS), you may want to have the fastest server with the fastest CPUs in it. However, if you study this carefully, you would find out that the fastest CPUs alone are not the answer. It may be established with fast CPUs, memory, disk, and other components combined. Energy consumption at a data center can be considered in the same way. Networking connects IT equipment together and, by activating and deactivating according to needs, networking controls other IT resources (IT equipment). This energy efficient technology in networking should not be considered just for networking equipment but for all the IT resources and, consequently, the entire data center.

After talking with Wael and David, I pondered the following. My discussion with them reminds me of the argument about whether servers or networking are more important in computing. It is true that servers are the center of computing, and we tend to pay attention only to servers. However, networking ties together servers and other equipment, like storage, to create an IT system. Without fast, low-cost, and energy efficient networking equipment, even the fastest and most energy efficient servers could not deliver, alone, the most energy efficient computing for a data center. This means networking equipment needs to be energy efficient. On top of that, networking equipment should be able to allocate resources dynamically, as needed, and without delay, to ensure the energy efficiency of other IT resources.

When the IT energy efficiency discussion was in its infancy a few years ago, it was O.K. to discuss each category of IT equipment in a data center separately to see how much it consumed in order to control its consumption. As IT equipment is being integrated and coming to function as one system, it is time to consider the energy efficiency of the IT system as a whole.

There are a few more technologies for making networking equipment more energy efficient, but I will cover them sometime in the future.

Simon Tusha, CTO of Quality Technology Services (QTS), gave a talk on this subject. Incidentally, several months ago I visited QTS’s Silicon Valley site for a data center tour and a discussion of their view of market trends. I’ve also reported on their CEO’s presentation of their business.

By the way, QTS recently raised $25M and expanded by 70,000 sq. ft. and 15 MW. The company had received $150M from the same investor before.

Simon Tusha

Tusha presented a bunch of statistics and market trends in his talk. I’ll summarize some of the highlights.

First, the data center market is growing despite the current recession.

The market drivers of data center growth are in many areas, including online business and financial processing, online medical records, and several regulations (SOX and HIPAA, for example). Because of growing access to the Internet and data centers, data center operators suffer from a power and space shortage. This trend has been going on for some time, and no substantial remedies have been applied.

According to Tusha, those problems are manifested in a survey conducted by the Data Center Users Group. For example, 96% of current facilities are projected to reach capacity by 2011. Forty-four percent of them indicated power and cooling constraints.

Tusha then showed stats indicating that more and more companies are moving to colocations instead of building data centers.

This somewhat obscure slide (the quality of the picture, not the contents) shows leasing (blue) and own (red) volumes from 2009 to 2013. Clearly, leasing is increasing its percentage. In presentations elsewhere, the reason for that has been explained by: (1) the shortage or lack of capital to build a data center and (2) the time to market (it usually takes 18 to 24 months before a new data center is operational). In 2008, the financial crisis caused access to capital for data center construction to dry out, and the lack of data center space has persisted since then. The two reasons combined make it very hard for enterprises to develop their own to serve their needs. Instead, they want to go for colocation.

In addition, Tusha mentioned that anything under 30 MW does not make sense from the financial point of view. It costs approximately $9M to build out a 1MW capacity data center. But if you build a data center whose capacity is greater than 30 MW, the cost for 1MW goes down to $3.5M. (Rich Miller of Data Center Knowledge reported this same claim from another conference in more detail.

This, in general, goes with economies of scale for data centers.

Tusha shows the hierarchy of data centers in the following slide, which classifies the data center market into several groups. Each layer leases space to the layer immediately below it.

• First layer (30 MW or more): data center real estate developer
• Second layer (less than 3 MW, from left to right): colo/interconnect provider, service provider (hosted/cloud), and large enterprise
• Third layer (less than 1 MW, from colo/interconnect player): service provider and medium enterprise
• Fourth layer (less than 500 kW, from service provider): small enterprise

His analysis suggests that almost everyone should go to a colo provider instead of building their own data center. In order to beat competition, each colo provider should become bigger by acquisition. This fact sort of explains why data center M&A is very active, as reported elsewhere, including Equinix and Digital Realty Trust.

Andreas Zoll

Zoll started his talk by giving the current status of data centers, which was not surprising to those of us who keep track of their trends and challenges. These are some of his key points:

• Continuing growth of densities (for computing, power, and cooling)
• Cooling requirements and focus
• High capex and opex
• Right-sizing hard

Container data centers could address most of those trends and challenges. He made these points:

• Rapid scalability
• Capex in line with actual demand
• Geographically agnostic
• Higher efficiency
• Repeatable
• Flexible

One of his slides shows the history of container-based data centers:

Container-based data center progression

He defined three generations of container data centers.

Generation 1:

• Box containing servers
• Vendor lock-in
• Niche audience only

Generation 2:

• Variety of cooling options
• Vendor-neutral configurations
• Customizable
• Not self-contained but requiring several vendors

Generation 3 (next generation that is not yet on the market):

• Purpose-built
• Open architecture
• Fully integrated
• Tailored for redundancy
• Maintainable
• Location agnostic
• Automated
• Internationally plug-and-play

Some people asked him to elaborate on his last point about the second generation and full integration in the third generation. The current generation of container comes either with IT equipment that has some power and cooling inlets or with power and cooling equipment. HP (IT equipment) teamed up with Active Power (power) to provide a whole solution. Dell’s version comes as a double-decker: the bottom is IT equipment storage and the top is a power and cooling box.

Regardless of configuration, power and cooling should be available as a modularized unit to hook the container up to the building. This reminds me of the fourth generation data center proposed by Microsoft.

What you can do with each container is limited. You still have to hook your container up to power and cooling before you put it into operation. So the entire data center should be designed to provide modularized power and cooling. How soon can you do that? It is a good idea, but I wonder how doable it is now.

While people were buzzing with this question, a VP came on stage and said i/o Data Centers would announce their latest container-based solution in two to three weeks. I suppose it will be the third generation. They planned the session well. They well knew that people would ask what "fully integrated” meant and that they would announce their new solution. That was very clever. We will see if their new solution is as good as their session planning.

After the departure of Andrew Fanara from EPA, a new person is in charge of Energy Star for data centers. She is Una Song.

She gave a talk on EPA’s efforts on Energy Star at the recent DatacenterDynamics in San Francisco.

Una Song

It was a good summary of Energy Star for data centers. Here’s a set of videos from her presentation. There was a little surprise during her talk. One of the participants started questioning why power generated with renewable energies is not counted in the computation of PUE. His claim is that power generated by clean energy should be given some credit. In an extreme argument, he says PUE could be 50, yet all the power fueling the data center might be from renewables. The absence of GHG emissions in that case should be commended rather than the bad PUE frowned upon. That interaction is captured toward the end of videos 2 and 3. This reminds me of the metric (called Data Center Performance per Energy, or DPPE) proposed by Japan’s Green IT Promotion Council.

This metric takes the use of renewables into account. To refresh your memory, here is the definition of DPPE:

DPPE =

(IT equipment utilization * IT equipment ability)

/ (total power consumed – green power used)

DPPE was also mentioned in another session. That will come later.

1. Video of Una Song’s presentation on Energy Star for data centers-Video1
2. Video of Una Song’s presentation on Energy Star for data centers-Video2
3. Video of Una Song’s presentation on Energy Star for data centers-Video3
4. Video of Una Song’s presentation on Energy Star for data centers-Video4

As in any normal data center operation, cloud computing must forecast the level of demand to prepare enough IT resources. Because it is more difficult in cloud computing to predict future loads, the cloud operator may be inclined to prepare a large number of servers, just in case. But the predicted peak may never appear. Unless the operator dynamically turns the servers on and off, some of them may run idle, consuming power but doing no useful work. Even when they are turned off, they still take up space. If that is the case, is cloud computing energy efficient?

On the other hand, what if the operator does not prepare enough IT resources, but large demands come in? He can either turn them away or reroute them to other data centers (clouds) by load balancing. In theory, this works fine, but how likely is it to succeed? I am not sure.

What about private clouds? Before cloud computing (private), each department in an organization tended to have a separate set of dedicated servers because each department had a separate IT budget. With the advent of private clouds, it became possible to mix and match IT loads from different departments on the same server, cutting power consumption. Again, it is necessary to forecast enough IT resources to prepare for the loads. The same argument holds in this case as well.

Energy efficiency by cloud computing makes sense when it is possible to predict the necessary resources well. But it is certainly better than the current practice.

I will interview Marten Mickos, CEO of Eucalyptus, in two weeks. Eucalyptus provides tools to create private clouds, as opposed to public clouds. I will pick his brain to see if my way of thinking is correct.

Although its conferences are international in nature, each conference has some local flavor. The upcoming one in San Francisco has a number of movers and shakers in the data center space, in addition to national and international participants. See the program here.

I plan to attend the conference as press and report on some of the sessions in my coming blog posts. If you are in the San Francisco Bay Area, this is a good conference to attend. And if you do attend, please find me and we can chat about data center issues. I am looking forward to seeing you there.

At ConnectivityWeek several weeks ago, a representative from the Federal Communications Commissions (FCC) explained that their broadband plan includes a section on energy and the environment. More specifically, it considers smart grid from the communications perspective. The entire plan can be downloaded from here.

Chapter 12 (pp. 245–262) of the plan contains a section on how communications and broadband can help smart grid.

The entire plan is 376 pages long; I will only discuss chapter 12, which relates to the theme of my blog.

Chapter 12 has four subsections:

• Section 12.1 discusses how smart grid is implemented and reinforced with the application of broadband communications technologies. It gives a good overview of smart grid 101. If you want to get a glimpse of what smart grid is, especially from the communications perspective, this is a good section to read.
  

1. Enhancing the communications infrastructure.
2. Using commercial service providers (who are often considered not reliable enough for this crucial infrastructure).
3. Strengthening communications security.
4. Opening up the 700 MHz Public Safety Band for utilities to use.
5. Further opening up the Public Safety spectrum for smart grid.
6. Studying the requirements of utilities for smart grid.

• Section 12.2 covers home and building aspects of smart grid from the communications point of view. Smart meters installed at homes and other buildings would make power consumption information readily available for utilities and consumers. This section recommends:

1. Mandating utilities to provide consumers’ use data and maintain the privacy of this data at the state level.
2. Doing the same for the Federal Energy Regulatory Commission.
3. Doing the same for the Department of Energy.
4. That the Rural Utilities Service provide loans to rural utilities to implement smart grid in rural areas of the United States.

• Section 12.3 considers ICT as a whole in regard to power consumption and greenhouse gas emissions. It discusses PCs as well as data centers. This section recommends that:

1. The FCC start working on the communication industry’s energy efficiency and environmental impacts.
2. The federal government start working on the energy efficiency of data centers.

• Section 12.4 is about smart transportation. Since this is a report on broadband, rather than electric vehicles (EVs), this section is about how smart communications technologies can help reduce gas consumption by providing appropriate information, such as how to reroute to avoid traffic congestion.

If you have time to read the entire report, it would certainly help you grasp what the FCC is considering in the way of improving the nation’s communications infrastructure. But if you are daunted by the length, chapter 12 alone would give you a good glimpse of how ICT could benefit smart grid. I have not seen any article or report discussing smart grid from the communications perspective before.

By the way, my friend in Saratoga, CA (which is pretty affluent), still complains of the lack of broadband services to his city. Saratoga is the community next to mine, but we have a choice of DSL and cable. My friend often tweets and expresses his opinion on the sad status of broadband services in the U.S. I just cannot imagine that that status is caused by the lack of technologies. It must be something to do with policies and politics. But it must be frustrating not to have broadband services at home.

I can relate to the frustration. Incidentally, my wireless router is acting up and drops the Internet connection sporadically. When I am not at home to fix it (cycle the power?), my wife gets angrier by the day. It is about time for me to consider a new router with 802.11n, which is somewhat pricey compared with the one with 802.11g. But again, this is a good excuse for me to purchase one.

Here’s the brochure for the 2010 Data Center Efficiency Summit.

First, the statistics. Who has responsibility for the power for data centers? Viridity’s data, as shown here, indicate a growing percentage of IT people becoming involved and sharing responsibilities for power at data centers. It has been said many times in the past that the power consumer (IT) should get or at least share responsibility for power at data centers with facilities personnel. It appears that message is finally sinking in. It is simple logic that unless consumers are responsible for power, they never pay attention to how much power they consume or how efficient their equipment is. This is a good trend.

Second, problems with data centers. As the following chart indicates, the first and second problems are power cost and lack of space. If we combine the third (lack of power) with the power cost, the power-related concerns are 35%, by far the largest piece.

Third, how data are measured. Viridity said that in a typical survey, the usual breakdown was one-third of respondents don’t know how their data are measured, one-third use sensors, and one-third use nameplate (derating) information. Viridity reported that measuring with software is increasing, as shown in the pie chart below.

The collection method really depends upon what data you are trying to collect. There are some data you can collect without sensors, but other types of data require external sensors. Power consumption on servers is readily measurable via SNMP, and it can be used for automatic discovery of IT equipment. If you focus only on power consumption, you probably do not need external sensors. However, data like pressure, temperature, and humidity cannot be measured without external sensors.

I started to list functions for metering and monitoring services by vendors. See here for those functions. The Viridity demo during the webinar talked only about asset management and power monitoring. I cannot judge its functions fairly since I do not know the software’s other functions. I only focus on their asset management and power monitoring functions. You can see their demo at their website.

They attempt to solve the problem of allocating the necessary minimum power to satisfy the IT load. With the good integration of its CAD function and power consumption information, you can create, manage, and maintain your data centers. The information available via their dashboard includes:

• The floor layout with rack placement information
• Power consumption at each rack and each server
• Each server utilization
• Availability of total power and available power at each rack

The software tool allows you to change the view from the entire data center to each server. More functions could be added to enhance their tools, but some sensor integration would be necessary.

Credits: 3 above charts are by Viridity

I am not moving my old posts on smart grid or renewables to the new area. However, for your convenience, I list the recent posts relating to this area:

• EV Panel Session at the Fujitsu Smart Grid Symposium
• Smart Grid VC Perspectives
• Fujitsu’s Smart Grid Symposium
• Interview with Kurt Yeager of Galvin Electricity Initiative
• Interview with John McDonald, Chair of Smart Grid Interoperability Panel Governing Board
• Interviews at ConnectivityWeek
• Declaration of Energy Independence and Consumers' Bill of Rights: Power to the People
• Japan Shows Her Face in Smart Grid Efforts
• When Vint Cerf Speaks, Everyone Listens
• More on Japan in New Mexico’s Smart Grid Market
• Solar Market in Japan

Currently, electricity enters a data center in the form of alternating current and goes through several conversions between AC and DC. When the alternating current reaches IT equipment, it is converted to direct current for operating that equipment. If we can convert the incoming AC power to DC once, at the entry point, it would be much more energy efficient.

In 2007, NTT Facilities participated in a workshop to test the energy efficiency gained by going with DC power. Recently, I met with Keizo Hoshijima, general manager of NTT Facilities’ U.S. office.

Keizo Hoshijima

We covered other business first, and when we talked briefly about DC power distribution at data centers, Hoshijima explained that, unlike the U.S., both Europe and Japan are ready to adopt DC for power distribution in data centers. In the U.S., NTT Facilities is working to lobby IEC to adopt DC power distribution. I found out that Hoshijima works with Dennis Symanski at EPRI on this matter. Symanski was on my panel at Nordic Green. 

Now I plan to visit Symanski for more details on this subject.

(Remember John McDonald’s comment about colocating relevant groups to encourage discussions for standards?)

Information about the ZigBee Alliance is here:

The ZigBee Alliance is an association of companies working together to enable reliable, cost-effective, low-power, wirelessly networked, monitoring and control products based on an open global standard.

The goal of the ZigBee Alliance is to provide the consumer with ultimate flexibility, mobility, and ease of use by building wireless intelligence and capabilities into everyday devices. ZigBee technology will be embedded in a wide range of products and applications across consumer, commercial, industrial and government markets worldwide. For the first time, companies will have a standards-based wireless platform optimized for the unique needs of remote monitoring and control applications, including simplicity, reliability, low-cost and low-power.

In addition, ZigBee is being considered as a communications protocol for home and field area networking for smart grid. Others include Wi-Fi, 6LowPAN, Z-Wave and HomePlug.

The following is an edited summary of my interview with Heile.

Bob Heile

Question: In your earlier presentation, you did not mention data centers as a potential application area.

Answer: You need to understand that ZigBee is trying to provide a communications protocol/platform to support ubiquitous connections among many objects anywhere. That would generate a lot of information and data. We are concerned with insuring that the needed data is collected and delivered not with how those data or information is managed.

Q: Several communications protocols are being discussed for smart grid. I stopped by the Itron booth (smart meter company, link to bob-heilie-by-zen-kishimoto-1.doc), and found out that Itron and other leading smart meter vendors have embedded ZigBee in their communications. Considering the large number of meter deployments and the fact that most utilities consider a smart meter a gateway [according to Greentech Media, 56% of utilities think that], ZigBee appears to be becoming the communications protocol of choice for smart grid. Is the competition over?

A: As you pointed out, ZigBee has been embedded or will be embedded into 40M-plus smart meters. We are about to cross the chasm [as in Geoffrey Moore’s Crossing the Chasm model]. It is, however, important to work with core wired and wireless communications protocols, such as Ethernet, Wi-Fi, and WiMax (802.11.6). These protocols will never go away, and it is very important for ZigBee to work with those protocols to establish ubiquitous connectivity to implement the "Internet of things.”

Q: So devices in the future will have a chip to support multiple communications protocols?

A: It has been already done though not always in a single chip just yet. Think of a Wi-Fi access point. It supports 802.11 a, b, g, and n, which are very different radios but the result seems seamless.

Q: You are currently implementing an IP stack. That means that meters with a ZigBee chip now need to communicate with a collection/aggregation node that speaks ZigBee on a pole nearby, for example. If so, who are those vendors?

A: Not really, it is just a different way of providing the same networking functionality being installed today in the HAN but based on IETF RFCs

Q: Once the IP stack is implemented, the meters should be able to communicate with almost anything, as most objects speak IP?

A: They can do that today. An IETF oriented solution is a familiar architecture and is preferred by some but as with anything it is not black and white. Some things get simpler and some more complex.

Q: What happened to UWB (http://en.wikipedia.org/wiki/Ultra-wideband)? I thought that was a cool technology. Was it killed because two camps with different ideas could not merge them?

A: From the beginning, UWB had a problem in commercialization as it had a problem from the global regulatory point of view. [Note: Allowed spectrum range was one of them.] A protocol like 802.11n can support up to 200 Mbps and has no regulatory issue, as it uses the ISM band [free spectrum]. Given the deadlock in 802.15.3a delaying getting anything to market, the global regulatory environment, and the resultant availability of 11n, UWB did not have a chance as a high data communications radio technology. Even if the groups had agreed it is not likely the outcome would have been any different.

Q: You talked about the horizontal expansion in application areas of ZigBee. However, vertically or technologically, are more feature enhancements to it necessary?

A: For the areas other than ZigBee Smart Energy, I think our basic core stack development is done with the version of the ZigBee specification that was published in 2007. We will make enhancements periodically as we gain more knowledge from the field and occasionally add support for new public profiles. The biggest area of new development is the profiles

Q: What is the battery life of an object with a ZigBee chip?

A: For example, the battery life of a light switch is easily 20 years. In general, it outlasts the owner of those devices.

Q: In your talk, you mentioned that about 30% of ZigBee Alliance members are in Asia. Can you describe what they are doing with ZigBee?

A: Asia is very diverse, but let me mention the five most active countries: Korea, Japan, China, Australia, and India. Korea is everything wireless. Japan uses it for disaster recovery and smart grid. China looks at the "Internet of things” as one of its strategic plans. Australia uses it for tackling drought and energy efficiency via smart grid. Finally, India, using its software development power, provides solutions with ZigBee for others.

Arch Rock is not alone. Many leading vendors have embedded a ZigBee chip into their smart meters. Such a vendor, Itron, had a booth at the recent ConnectivityWeek 2010 show.

Itron showed meters with ZigBee chips in them at ConnectivityWeek

At the same conference, Bob Heile, chairman of the ZigBee Alliance, presented that group’s most recent developments.


Bob Heile

I videotaped his talk (40 minutes long) in the following videos.

Heile made several noteworthy points:

• The ZigBee Alliance is not a vendor organization but an international nonprofit organization to promote the ZigBee specification (low-power wireless and mesh networking). It consists of 360 companies worldwide.
• The membership spread is North America 40%, Europe 30%, and Asia 30%.
• The applicable markets include Consumer Electronics, Energy Management and Efficiency, Health Care, Home Automation, Telecommunication Services, Building Automation, and Industrial Automation.
• As for smart grid and energy concerns, 40million ZigBee smart meters have been contracted so far and are being installed at a rate of of over 20,000 meters/day.
• Most notably for me, working at the intersection of IT and energy, is their announcement of ZigBee Green Power, which obtains the necessary energy to fuel a ZigBee-supported device without wires or batteries. One of the concerns, from the capacity and e-waste points of view, about using a wireless sensor is its power source, a battery. ZigBee Green Power appears to solve this problem.

I have talked to more than half a dozen vendors in this space. See my original post for their identities. Although they are all in the same monitoring and metering space, each company has a slightly different angle. Some measure via their own sensors and aggregate the data for display. Others do not gather data by themselves but exploit the data from other sources. So the functions can be roughly classified into three categories: measuring (via sensor), aggregating, and analysis and display.

Also, in addition to measuring power consumption, different approaches can be taken to common functions like:

• alarm management
• asset management
• capacity analysis
• efficiency analysis
• air control automation

In this panel, I would like to discuss the following so that a general audience can understand the needs of metering, learn which technologies are state-of-the-art, and assess the minimally necessary functions of metering.

We would discuss the following:

• Why do data centers require monitoring and metering?
• What is a typical architecture for measuring, aggregation, analysis, and display?
• What are the minimally required functions of metering?
• What kind of standards should be defined for metering? What type of data should be collected? Granularity of data? Frequency of collection? Data formats (like XML)? Display formats?
• What extensions, such as e-waste and water, should be considered for existing metering?

An ideal panel will consist of three or four customers of vendors, one researcher in the space, and a regulatory representative, namely EPA. As of this writing, I have received positive responses, but this is just the beginning. I will report how this goes in the coming days.

This data center, called SC1, is the first in Silicon Valley for Dupont Fabros. It is approximately360,000square feet,and it will have a total power capacity of 26megawatts. Here’s a picture of the revitalized Dupont Fabros’ SC1 data center construction site.

To answer these questions and others, the latest Fujitsu Labs of America technology symposium had a panel on EVs. The panel consisted of:

Moderator:

• Mike Liebhold, Senior Researcher, Distinguished Fellow, Institute for the Future
  

Panelists:

• Sven Thesen, Head of California Business Development, Better Place
  
• John Suh, General Motors, Advanced Technology Silicon Valley Office
  
• Serdar Uckun, Manager, Embedded Reasoning Area, Palo Alto Research Center
  
• Jay Friedland, Legislative Director, Plug-in America
  

I post the presentation by Better Place only because Thesen showed an interesting video. It was about their experiment in Tokyo using an EV powered by their battery technology as a taxi.

Better Place leases batteries and keeps them at their charging (or should I say exchange) stations. An EV owner does not own a battery but leases one from Better Place. When the battery runs very low on power, the owner drives the car to a nearby exchange station. It takes about a minute to replace a dead battery with a fully charged one.

The company originated in Israel and is doing well there. When I first heard about it, I thought their success depended on the standardization of this practice, including the removable battery architecture. Unless the majority of EVs adopt this practice, it will not work. It is a capital-intensive business as well, because you need to keep fully charged batteries available and because you need to build battery replacement stations all over town. Small countries like Israel and Japan may find this business attractive, but I wonder if the same thing would work in the U.S., especially in the west, where the territory for the replacement stations to cover is pretty wide and people drive longer distances.

John Suh of GM posed six questions in addition to the two issues I mentioned above: 

1. Which is better for the environment, EVs or fuel-efficient gasoline-based cars?
2. Starting in 2011, if we started with EVs, how long would it take to replace all the cars in the U.S.?
3. What is the difference between plug-in logic hybrids, extended range EVs, and hybrid logic vehicles?
4. Where are the two most suitable locations for charging stations?
5. What are the most inexpensive and effective technologies for reducing the death rate and increasing energy efficiency per mile?
6. What do food dieting and energy policy have in common?

After the initial monologue by each panelist, Liebhold’s specific questions to GM’s Suh were noteworthy. They concerned:

• battery compatibility
• plug compatibility
• working with utilities to control capacity

GM has been agnostic about battery types, and besides, it is hard to standardize batteries on EVs. It might make sense to standardize by class of vehicles, though. Plug compatibility is guaranteed by observing the Society of Automotive Engineers’ (SAE) J1772 standard. Regarding capacity planning, GM works with utilities though the Electric Power Research Institute (EPRI). Also through EPRI, it works with the Department of Energy. The onboard OnStar system communicates with the power grid system to send and receive information regarding its demands for power and the price for that power.

Both Friedland and Thesen countered the negative comments on EVs by saying that perception is beating reality. Their arguments hold for some cases, but mass adoption needs more time.

Bjarne Stroustrup

He is the creator of the C++ language.

C++ was conceived from C and Simula.

I would not mind talking about C++ and other programming languages for the next ten blog posts, but this is a blog about energy efficiency, green IT, cloud computing, and smart grid. It is Fujitsu’s fault for choosing him as a keynote speaker, which brought my mind back to pure IT.

In any event, as more and more objects are equipped with the ability to send messages back and forth, the "Internet of things” is everywhere.

The theme of Stroustrup’s presentation was how to bridge the gap between applications and hardware using C++ and C++0x (the latest standard version), as shown here.

Bridge the gap between applications and hardware using C++ and C++0x.

After talking about what is not an infrastructure, he listed the necessary conditions for being an infrastructure.

What is required of infrastructure?

He also mentioned the difficulty in programming in the gap, which is supposed to connect applications and hardware. He said the interaction between the lowest layer of the infrastructure and hardware is very complex. For example, Intel has about 200 software engineers working on a compiler to produce appropriate code to link the lowest layer and hardware.

Programming in the gap is difficult.

So the question is, how do you implement such "things” in an efficient way? And the answer is to use C++. See the next picture.

Why C++?

Stroustrup did not say so explicitly, but he was talking about embedded systems. In the world of embedded systems, unlike the world of the Web, C and C++ still dominate. C can be very low-level language and interact well with hardware without requiring a lot of such resource. C++ is replacing C because C++ improves many of C’s problems, but I am not going there in this blog.

His conclusion as to how to bridge applications and hardware is given in the following slide.

The panel consisted of:

Moderator

• Eric Wesoff, Senior Analyst, Greentech Media

Panelists

• Steve Vassallo, Venture Partner, Foundation Capital
• Tammi Smorynski, Director, Intel Capital
• Josh Becker, General Partner, New Cycle Capital

From left to right: Wesoff, Vassallo, Smorynski, and Becker

What is the killer application for smart grid? This question is asked at every smart grid conference I go to. Most times, the answer is that after automated metering infrastructure (if that is counted as an application rather than an infrastructure), it is demand and response (DR). DR is based on a simple idea that shaves the peak demand and adjusts the entire demand to stay within the supply. If you press more, the usual answer is that it’s an application you haven’t considered yet in your wildest dreams. I am not sure whether the home area network (HAN) applications are considered killer, but there are many companies in this space.

This session covered a lot of ground, but one thing that caught my attention is the promising market for smart grid. A large number of startups and established companies have entered the HAN space. Wesoff said the number is tens if not hundreds. This is easy to explain. When you look at the entire power grid, the transmission side involves a large capital investment and specific expertise in high voltage transmission. It is not easy to obtain either one. Even on the distribution side, it is still capital-intensive and requires expertise in such areas as renewable power generation and managing/integrating the power generated by renewable energies.

On the other side, automated metering infrastructure (AMI) and field area network (FAN) are straightforward ICT applications requiring relatively little capital and power expertise to implement. HAN appears to be very similar to those two. Moreover, it seems to be easier to work in the HAN sector. We are familiar with home appliances and can easily test them with monitoring and metering. This also means the barrier to entry is not very high.

Incidentally, Vassallo’s firm invested in Silver Spring Networks (AMI/FAN) and EnerNoc (DR), which is public.

Smorynski of Intel Capital was skeptical about the HAN market growing into a viable market segment within smart grid.

Her points were two:

1. The average person does not care about day-to-day or hour-by-hour power consumption as long as he or she is reasonably happy with his or her electricity bills. The average person sets the energy efficient setting once and forgets about it.

2. In a typical home, HVAC (air conditioner and central heating) consumes the majority of power (some 60%). Also, people may be sensitive to the use of lights. But after those two, there is not much a consumer can do.

So the home energy management (HEM) sector does not seem to attract much attention from the average Joe or Jane. So far, DR has been implemented for commercial and industrial markets. The HEM market may move when DR includes the majority of residential power consumption data. Even when that is the case, will the average person at home stare at his or her energy consumption display every ten to fifteen minutes? I know myself. If I get that kind of gadget, I may stare at it quite often for the first two or three days and then forget about it.

It remains to be seen if this analysis is right. One thing is true: no simple and easy thing can win the market.

As you know, smart grid is at the intersection of power, telecom, and IT, and that means a lot of subjects to cover. Designing a one-day conference with so many subjects takes a lot of effort to coordinate. Fujitsu did a good job of packing valuable sessions between 9a.m. and 6 p.m. on June 9. The company did not charge for attendance. On top of that, it provided breakfast, lunch, and hors d'oeuvres with drinks. But that’s not why I’m giving it high marks.

Hitoshi Matsumoto, president and CEO of Fujitsu Laboratories of America, opened the symposium.

Hitoshi Matsumoto

Eric Wesoff, senior analyst, Greentech Media, delivered the first keynote.

Eric Wesoff

Wesoff’s talk can be described as smart grid 101. It was a good summary of what smart grid is in layman’s terms.

The following is a list of smart grid topics covered during the symposium. I will blog on some of the sessions in the coming days.

• utilities’ perspectives
• Internet of things
• C++
• network security
• consumers’ perspectives
• electric vehicles
• VC perspectives

The C++ session was by Bjarne Stroustrup, Distinguished Professor and College of Engineering Endowed Chair in Computer Science at Texas A&M. I used to follow his work and studied C++ intensively. His talk was very different from the rest. I will definitely blog on his session in the coming days and explain why his work is relevant to smart grid.

Finally, the symposium agenda is attached.

• Dell’s Double Decker Container Data Center Building Block
• Container Based Data Centers (DCs)
• Container-Based Data Center
• Nebula Cloud Project-in-a-Box by NASA Ames
• Internet Archive
• A Container as a Building Block for Data Centers

Containerized data centers have received most of the attention so far, but modularization is a definite trend in data center design and construction. John Diamond of DAS Associates wrote an article about this trend, "Modularized Facility Infrastructure Can Control Scaleup Costs,” for Mission Critical Magazine. 

When Microsoft announced an idea for the fourth generation data center The Fourth Generation Data Center, it showed that future data centers could be built from a number of well-defined smaller pieces. Diamond’s article reinforces that vision.

It is often pointed out that IT and facilities need to share information and collaborate to make data centers run flawlessly and in an energy-efficient manner. Diamond first talked about the innovation gap between IT and facilities.

In just one year, the technology community identified a risk to reliability and recovery and implemented a fix that was dramatically more efficient, while the facilities went unchanged and would remain so for the next 15 years.

Diamond argues that as virtualization has freed application software from specific hardware for its operation so that it can run anywhere, each data center no longer needs to be constructed to the rigid requirements of Tier IV. This leads us to modularized designs.

Today in the cloud environment and with the growth of network computing, new topologies have eliminated the need for large hardened infrastructure by applying network topologies to reroute data traffic to another data processing plant. It is this kind of dynamic flexibility that has allowed the mega data centers to minimize one-of-a-kind designs and custom building and take advantage of modularized construction much like the utility merchant plants.

With such a modularized design, almost all of the components at a data center are modularized, including generators, prefabricated UPS rooms, and switch gear modules. This definitely leads to less-expensive components and speedier construction using those components than is possible for custom-built data centers.

Beyond container-based data centers, a data center can be built using off-site construction. With the use of building information modeling (BMI), part of a data center can be built away from the operating site and later integrated into the operating structure. The pharmaceutical and textile industries have already done this. It is finally sinking into the construction of data centers. Modularization is here to stay.

The off-site construction industry has assembled full raised floor environments and drop shipped them to warehouse-style data centers developed since 2000. In less than the time it takes many design firms to program a data center design, a full data center can be built and shipped to any location. Complete with racks, cabling, above or below floor cooling topologies, networking, controls, security and fire safety all integrated into the delivered package, requiring simply the connections for power, cooling, and communications.

The 2010 Energy Summit is a signature event of both the Silicon Valley Leadership Group and the Stanford University Precourt Energy Efficiency Center, attracting a broad range of executives and representatives from influential Silicon Valley companies and organizations. Last year’s Summit drew a record crowd of more than 500attendees, and we anticipate reaching capacity well before the event this year as well.

The 2010 Energy Summit is an essential event for members of the Silicon Valley business community. Practical and inspirational, it serves as a manual for sustainable business by combining current best practices with a guide to upcoming technologies and government regulation regarding energy.

During the one-day event, participants learn thatsustainable energy use is about more than feel-good environmentalism; it is about planning for growth in a market of diminishing resources and changing stakeholder priorities.

As the name suggests, there are a lot of sessions and panels about energy matters. For this blog’s specific interests, there are plenty of smart grid and energy efficiency sessions.

Some smart grid sessions:

• Smart Grid: Technology and Feedback
• Your New Utility Bill: Where Is It Heading? Peak-Day, Real-Time Pricing, and Direct Access
• Smart Grid: Consumer Feedback and Response

An energy efficiency session:

• Energy Efficiency: Management Strategies

AltaTerra will publish a report, in conjunction with Silicon Valley Leadership Group, summarizing this summit, and some of us, including myself, will attend it. See you at the summit.

The following is an edited summary of my interview with Yeager.

Question: What does the Galvin Electricity Initiative do?

Answer: I was president of the Electric Power Research Institute (EPRI), and when I retired, Bob Galvin, the former chairman of Motorola and a member of the advisory group for EPRI, suggested that I join him to create an organization to encourage innovation of the power industry, and I joined him at the Galvin Electricity Initiative. The Galvin Electricity Initiative is leading a campaign to transform the way we generate, deliver, and use electricity.

Q: In your earlier speech, you implied a declaration of independence from monopolistic energy suppliers. Isn’t the microgrid concept right on target for your vision?

A: Complete independence from the power grid with microgrids is not practical unless we are talking about developing countries. What I am saying is that there should be a good balance between distributed generation and the power supplied by utilities and consumers should determine that balance. My vision is to open the door for consumer empowerment and innovation in the area of power generation, delivery, and use.

Q: When people talk about microgrids, they often mention that Denmark is full of microgrids. Can we use Denmark as an example when we implement microgrids in the U.S.?

A: Denmark is a unique country, and its microgrid system cannot be copied for the U.S. The philosophy behind microgrids is to optimize the distribution grid beyond each substation to support the recipients of power, whether they are residential, commercial, or industrial customers. Integration is key for this.

Q: Is microgrid implemented for each house or for a community?

A:, Microgrids primarily are for communities, but as technologies mature, in some cases they may be good for each residence.

Q: What is preventing microgrids from proliferating in the U.S.? Is it because of technology, money, or policy?

A: We have technologies and money. The public money necessary to kick-start this is not big. The regulatory climate maintains monopolies, noninnovation, and the status quo. The concept of decoupling should work to convince utilities to serve consumers well, since their total power sales and benefits are decoupled in states like California. So utilities should be evaluated for quality of service but not quantity of service. Smart meters are being deployed, but they are a quick and dirty fix to the current power grid. They may actually raise power bills without providing any benefits to consumers.

Q: Microgrids or distributed generation does not imply only renewable energies. Am I right?

A: Yes, you are right. It could be natural gas–based turbines generating the power. It could be fuel cells and combined heat/power generation.

Q: What about solar, wind, or fuel cells for microgrids?

A: Currently, most renewable power generation is by wind, but that will not last long. I believe that distributed photovoltaic solar power which converts buildings from power pigs to power plants in microgrids will be the major contributor. Fuel cells can be strategically promising, but their use is not cost- or environmentally effective yet. When some carbon regulations are put in place, those may provide an incentive to push fuel cells.

Q: In Japan, electricity and gas are provided by two different kinds of companies, namely, an electricity company and a gas company. Electricity companies are pushing all-electricity. To counter that, gas companies are trying to sell fuel cell systems based on natural gas. What about the U.S.?

A: As the economy gets better, the cost of natural gas will rise again, and it may become cost-prohibitive to do the same here. Also, fuels cells operate on hydrogen, which if derived from natural gas rather than renewable energy, wastes energy and carbon.

Q: We talked to Joe Sugg of Santa Clara University about their microgrid implementation, which is to start this summer. Is there a similar project in the San Francisco Bay Area? [Someone at the conference told me that UC San Diego is working on a campuswide microgrid.]

A: Santa Clara University asked Galvin for some advice. Also, there is a microgrid development being planned for Moffett Field area. Marin County is by far the most advanced in terms of renewable energy. However, the local utility is trying to discourage them from making progress on that front. [Incidentally, there is a proposition on the ballot for the June primary election.]

After this interview, Yeager had yet another presentation, this one called "Integrating Renewables into Grid.” It was a somewhat higher-level summary of my interview. It is also a good summary of the status of microgrids and renewable energies and of his vision for smart grid. The combined length of the two videos is a little more than ten minutes. They are worth your time.

• Kurt-Yeager-video-1
• Kurt-Yeager-video-2

John McDonald

At the beginning of the conference, it was reported that SGIP, which NIST set up to discuss the many technology areas needing standards, now consists of 600 companies and other organizations and more than 1,600 participants. I am one of those participants.

I am not the only one who has been wondering how such a large organization could be run efficiently. To run an organization full of strongly opinionated technology experts with telecom, IT, and power backgrounds, the Chair should possess a good technical background in all three areas and exercise strong leadership (an iron hand) with a large voice. I was not sure if such a person existed. My image for such a leader was completely betrayed when I saw John McDonald. He is a soft-spoken gentleman with the air of an executive. However, the first five minutes revealed his strong technical background and depth.

Yes, I had read his bio prior to the meeting, but what was on the paper had not sunk in with me by the time I saw him. Here’s an edited summary of my interview with him.

Question: How did you get to become Chair of the NIST SGIP Governing Board?

Answer: In addition to one member from each of the 22 stakeholder categories, 3 at-large members were elected last November for the Governing Board. In December, the Governing Board met for its first meeting at NIST. I was elected as one of the at-large members, and then I was elected Chair of the Governing Board in January.

Q: How do you run such a large organization?

A: The NIST SGIP Governing Board Chair deals with a much smaller group than the 1,600 memebrs of the NIST SGIP. That is much easier to manage.

Q: How do you come to a consensus when you need to deal with opinionated technical experts in the three areas of power, telecom, and IT?

A: Being able to foster free discussion is necessary. I am a good listener and also provide knowledge. However, at some point the discussion needs to converge to a consensus, and the chair needs to encourage that. I have done standards work for 36 years and have a background in all three technologies.

Q: Open membership is a good thing, but isn’t it too easy to sign up?

A: Keeping the membership open is better for soliciting a wide scope of opinions.

Q: I guess not all the 1,600 people are active in SGIP activities. Don’t you want to tighten up the qualifications?

A: There is a voting member from each organization, and we reach out to those members. Those members in turn reach out to other members of their organization, which works fine.

Q: Is the membership only for U.S.-based organizations?

A: No, there is no such restriction.

Q: You wear a lot of hats. How do you manage your time to accomplish so much?

A: One of my roles as Chair is to be a spokesperson for SGIP. I am a spokesperson for IEEE and GE as well. I also teach and enjoy explaining complex matters in an easy way. I have published three books and 36 papers and articles. I like what I am doing, spending a lot of time on it, and my family members understand..

Q: What about dealing with international organizations? Sometimes different countries have their own agenda with their own technologies.

A: China and Japan tend to have their own standards for their respective countries. Products based on those standards will have difficulty being marketed outside of their countries. My way of doing business internationally is to spend more time with the people I deal with and develop personal relationships. For example, I am a good ping-pong player and got to know the head of a regional Chinese utility very well by playing ping-pong with him. He had been a member of China’s Olympic team!

Q: In the morning session, you said that distribution is very important. What are the problems with transmission? I can see capacity and integration with renewable energy sources as problems.

A: In addition, wide area control and protection can detect and isolate problems to avoid blackouts. Synchrophasors [also known as phasor measurement units (PMUs)] may be applied to provide more information regarding the health of the transmission system.

Q: Do you have any advice for a guy who came from IT and wants to master the power industry technologies?

A: I have two pieces of advice. First, I have written several papers on the subject of the need to integrate the operational infrastructure with the IT infrastructure that you may want to read. Second, you want to attend tutorials at relevant conferences, such as DistribuTECH  and IEEE PES.

McDonald’s recommended papers authored or co-authored by him:

1. "Realizing the Power of Data Marts,” IEEE Power and Energy Magazine, pp.57–66, May/June 2007. This article discusses the power system in some detail.

2. "Steering Smart Grid Success Requires a Common Understanding” This is a summary view of smart grid.

Now I see why McDonald was selected as Chair. He seems to have plenty on his plate with the NIST SGIP and related standards work on top of his IEEE activities, teaching, and his day job at GE. But he manages his time to accomplish a lot. The term for the Chair is two years, and I am not sure whether multiple terms are allowed. But whoever comes after him will have a hard time duplicating what he is accomplishing.

Two days later during the same conference, McDonald participated in the "Industry Working Together” panel. His speech was an even more concise and well-summarized version of what I had heard in person. See the two video clips taken from his speech during that panel session:

• How to work together for standards-1 (approximately four minutes) He was specifically asked how to bridge NIST SGIP and IEEE. His points were that collaboration is key to establishing standards, and co-locating relevant technology groups is the best way to enable collaboration. Many relevant yet independently created groups must reach a consensus to establish standards in the power space.
• How to work together for standards-2 (approximately three minutes) He was asked what the call for action should be. He had two messages. The first is to get involved rather than sit on the sidelines and watch what others do. Second, people in leadership should be open-minded about different views to encourage lively discussion without ties to their current employers. But at some point they need to lead the discussion to convergence into a consensus rather than to a divergence of arguments.

There is a reason for this. In a typical data center, IT power consumption is by servers and storage. Networking equipment has been a distant third because it was necessary to combat the power-guzzling servers first and then the storage devices.

Dave Berry of Broadcom made a good comment about energy efficiency technologies in the networking area. According to him, these technologies include:

• IEEE EEE (Energy Efficient Ethernet): This is a specification for a more-efficient Ethernet.
• WOL (Wake on LAN): A network message turns a computer on when it’s needed and off when it isn’t.
• Network I/O Virtualization with Unified Fabric: Fibre Channel can be carried on top of Ethernet.
• Network Controller I/O Offload: Offloads networking tasks from central CPUs to offloading engines.
• Network as a Platform for Dynamic Power Management

I agree with him that until now, energy efficiency for networking has not been discussed. When we have a handle on power consumption for both servers and storage devices, it will be time to work on the energy efficiency of networking equipment. I wonder when EPA will develop an Energy Star specification for networking equipment.

1. John McDonald, chair of the governing board of the Smart Grid Interoperability Panel, general manager of marketing at General Electric, and many more titles

2. Bob Heile, chair of the ZigBee Alliance

3. Kurt Yeager, executive director of the Galvin Electricity Initiative

McDonald has a tremendous responsibility as leader of the Smart Grid Interoperability Panel (SGIP), which consists of 600 organizations and 1,600 members. On top of that, experts from the power, telecom, and IT space need to discuss standards for smart grid from their particular points of view, which seems to be quite difficult. I was very much interested in how McDonald runs SGIP for convergence, rather than divergence, of opinions.

I followed wireless mesh networking before and was interested in how ZigBee, UWB, and Bluetooth worked together. Among all the technologies, ZigBee appears to be leading the pack, and its application to smart grid is also working well. I wanted to know the answers to some questions that people don’t usually ask Heile.

Yeager’s organization talks a lot about microgrids, and I wanted to know the current status of this technology and its applicability within the context of smart grid. We had a good and frank conversation.

I will post my interviews with these experts in the coming days.

ConnectivityWeek gathered many experts and movers and shakers in the area of smart grid in one place, and I learned a lot. In addition, I had an opportunity to interview those influential people.

As the press release states, a Declaration of Energy Independence and Consumers' Bill of Rights were discussed. For those who were not in the "Building the Smart Grid Value Chain” session, it is not clear why. The term, Declaration of Energy Independence, was conceived during this panel discussion. The person responsible was Kurt Yeager, the executive director of the Galvin Electricity Initiative and former president of EPRI.

I have an excerpt of his speech in this video.

In his speech, he clearly articulated what smart grid is supposed to be and dismissed some ambiguous definitions and perceptions, such as smart meters being smart grid. The video is only four minutes long, but his eloquent speech should be quoted as the reason why smart grid is being developed and why we the consumers be given choices and control over the energy we use.

Looking from the U.S., Japan appears to be a virtual single entity, as if a single entity could represent the entire country of Japan. On the contrary, there are 120million people behind that perceived entity. In this conference, there was a surprise. There was a panelist from a Japanese organization that focuses on smart grid in Japan.

Seiji Morishima of NEDO was one of the panelists in the session entitled "Creating a Smart and Green Innovation Tsunami.” This panel had many participants, and I will summarize it in a forthcoming blog.

But I will focus on Morishima’s talk here.

Panel

Seiji Morishima (middle)

His talk was primarily about NEDO’s philosophy and plan of attack for developing smart grid in Japan. I covered their plan of attack, including working groups for their specific directions, in a past blog.

Morishima said that Japan was thinking of collaboration and cooperation with other countries, such as the U.S., to develop standards for smart grid that can be shared worldwide. However, not all Americans take his statement at face value. Some are skeptical about it. I heard such complaints at the conference. To date, Japan has pushed its own standards and not complied with world standards in such areas as cell phones. So I cannot blame skeptics. Japan’s cell phone market is saturated, but Japanese cell phone vendors cannot compete in the world market, mainly because their standards (Japanese) do not comply with the de facto standards in the rest of the world. So the Japanese government and vendors are changing their stance towards standardization. Hopefully, Morishima’s attendance is the first step towards working on world standards.

The Japanese folks I talked with about this issue told me that world standards are not the same as the U.S. standards that are often pushed as world standards. Oh, well. You be the judge.

Another complaint I hear often about Japan is the lack of published information. I sat down with Morishima at lunch and asked about this. He said it was not intentional. His team is very small and it is hard for them to publish information. To make the matter worse, even if they publish information, it is usually in Japanese. Western media representatives stationed in Japan rarely understand Japanese (with some exceptions), so the amount of published information in English is limited. I asked Morishima to give me more information on their efforts in smart grid. My advantage here is that I can work in both Japanese and English. I will keep publishing the information from Japan.

It is a nice, cozy suburban environment without any distractions (nothing else to do except for conducting research).

There we saw the state-of-the-art technologies being developed by NTT Facilities. We saw three research projects (only a subset of all the research) going on for data centers:

1. Earthquake simulator
2. Airflow simulator
3. Environment simulator

When I say a simulator, it is not a software simulator but a real system consisting of a lot of iron. Let me briefly describe each.

Earthquake simulator

Japan has a reputation as the center of earthquakes, especially in the Tokyo metropolitan area. Since Tokyo is the center of economy and politics, earthquake damages to its infrastructure would be catastrophic, including for IT and communications. Yet not many businesses leave Tokyo because of that. The second-best idea is to get ready for imminent earthquakes. What does that mean for data centers and the equipment in them?

Equipment, especially IT devices, is sensitive to shaking. For example, disk heads may crash if the equipment is shaken too hard. If there is a way to mitigate such shaking, the equipment may survive an earthquake of considerable size. But how do you prepare for that without actually shaking your racks full of sensitive IT equipment? As shown in the following picture, NTT Facilities developed an earthquake simulator.

It is a giant platform with a few arms that can move and shake the platform vertically and horizontally, as shown in the two pictures.

(Note: All the pictures in this blog are due to courtesy of NTT Facilities, Inc.)

The first picture shows the size of this simulator. The second picture shows some details from above, such as the vertical and horizontal arms.

NTT has offices throughout Japan, and each is equipped with meters to record earthquake information. In addition to the data recorded by Japan Meteorological Agency, it has its own recording of the major earthquake in the Kobe region, which was not known as a site for major earthquakes prior to 1995. Initially, NTT Facilities did not allow us to take pictures. I can say this: even if you could take pictures of this sophisticated simulator, it would be impossible to copy what they developed. It involves elaborate manufacturing technologies and precise software to repeat the exact earthquake waves.

Airflow simulator

The second tour was of a chamber for experimenting with the control of airflow in a data center. NTT Facilities has a product to contain a cold aisle, also known as aisle capping. Two aisles are set up, with and without capping, and airflow and temperatures at multiple locations are measured.

The picture shows an aisle without capping.

Environment simulator

The last tour was of a chamber for controlling the environmental parameters (temperature and moisture) to test CRAC units for outside. CRAC units are usually deployed in a pair, one inside and the other outside. The outside unit is affected a lot by outside environmental conditions. It would be nice to be able to change the parameters and test them in a simulated environment.

This picture shows this chamber, which can produce temperatures ranging from 14 °F to 122 °F. When we entered this chamber, it was set to 14 °F to simulate a typical temperature at Sapporo (the biggest northern city) in winter. We only stayed there for five minutes because it was too much.

This tour exemplifies the fact that Japan has a lot of technologies in progress and sometimes shows them to the public (as long as they speak Japanese) but is very shy in presenting them in English. I do not think it is because of their secrecy but the lack of language ability to do so. I will keep publishing interesting news.

Vint started with the history of the Internet and said that in spite of popular belief (to survive nuclear attacks), the Internet’s initial objective was to share computing resources. There were only four nodes then.

He then described how the Internet should change to accommodate the new situation when every object has an identity and can communicate with any other object. You can easily guess what the new requirements will be like: large IP address space (32 bits for IPv4 vs. 128 bits for IPv6), and authentication/encryption of each packet for secrecy and authentication. Another point that struck me was that adaptability will be important because those objects may move around to exploit many different communication venues. You cannot assume only a handful of ways for communication. He said that the most difficult problem will be error caused by misconfiguration, which is hard to detect. A good way to detect misconfiguration early would be a good technology to develop.

I do not summarize what he said here, but you may want to see his entire presentation. It is worth 60 minutes of your time.

• Vint-Cerf-keynote-1
• Vint-Cerf-keynote-2
• Vint-Cerf-keynote-3
• Vint-Cerf-keynote-4
• Vint-Cerf-keynote-5
  
• Vint-Cerf-keynote-6

Although the House of Representatives passed cap-and-trade last June, the bill has not moved on to become law. Aside from this legislation, the data center industry should start preparing for a set of regulations to curb power consumption and, thus, greenhouse gas emissions. Since these emissions cannot be measured directly, we need a proxy for them: power consumption. Actually, a panel discussion on GHG emissions and data centers may be a follow-on to the first panel idea.

There are two phases for data center operators to consider. The first is measuring to understand actual power consumption. Without knowing current consumption, we have no way to curb it by applying a set of energy efficiency measures and practices. This is the easier of the two phases, as curbing power consumption would reduce the cost of running data centers. Measuring requires an infrastructure, such as sensors and their accompanying software. Once such an infrastructure is in place, the second phase will be relatively easy to prepare.

The third possible panel, on GHG emissions and data center operations, will consist of one or both of the people mentioned above or someone else with similar knowledge who feels that this subject is urgent. Both Smith and Manos have a keen sense of urgency, as their data center operations spread internationally, including the United Kingdom, which has a very strict carbon reduction commitment.

Other panelists could include:

• A representative of one or two metering companies that can generate the GHG emission quivalent from power consumption.
• A sustainability expert who sees the entire corporate sustainability picture beyond data centers.
•  Regulatory officials, such as from EPA and DoE.

Questions to the panelists may include:

• How soon should we expect a set of regulations to control GHG emissions?
• How can we prepare for these regulations, including preparing a measurement infrastructure?
• How do we think about sustainability as we consider GHG emission control and beyond?

There are several ways to curb server power consumption in data centers:

1. Turn off unused or unnecessary servers.
2. Consolidate servers via virtualization.
3. Implement power management.
4. Refresh hardware for better energy efficiency.

The second and fourth options above have been given a lot of attention. Is there anything more we could do to reduce consumption?

Storage

After we got a handle on server energy efficiency, we started looking at storage devices, which consume a sizable amount of power, even though it’s less than servers use. This is reflected by EPA’s recent effort to establish an Energy Star rating system for storage. There are a few energy efficiency technologies for storage devices:

1. deduplication
2. MAID
3. thin provisioning

Anything else?

Networking (revised and added)

We have not seen any effort on a rating system for networking energy efficiency yet by EPA. But IEEE has started working on one. Some of the energy efficiency in this space are:

1. IEEE EEE (Energy Efficient Ethernet): This is a specification for a more-efficient Ethernet.
2. WOL (Wake on LAN): A network message turns a computer on when it’s needed and off when it isn’t.
3. Network I/O Virtualization with Unified Fabric: Fibre Channel can be carried on top of Ethernet.
4. Network Controller I/O Offload: Offloads networking tasks from central CPUs to offloading engines.

When we analyze power consumption at data centers, depending upon the data we quote, we find that cooling consumes 30% to 60% of the power. So we tend to curb the power consumed by cooling. But when we step back and think hard, we realize that the IT equipment is generating intense heat. There is a study that indicates that a 1 W IT saving equates to a 2.84 W facilities saving.

So another idea for a panel is to gather people who are in the server, storage, and networking space and discuss how the IT space can be made more energy efficient. One panelist could be a vendor or someone who installs data center equipment. Another panelist could be a designer or planner of IT infrastructure who might be able to quantify the saving. If the quantification is hard for the planner, we could bring in a researcher.

I might ask the panelists:

• Can you run your equipment under much higher temperature? If so, how high can it be?
• Is there any other way to curb (server, storage, or networking) power consumption further, independent of the facilities environment?
• Can you save more if you coordinate your server with storage or networking equipment?
• How can you curb power consumption if you coordinate your IT equipment with the facilities environment?
• How about DC interfaces to your IT equipment to support DC power distribution?

I have a few ideas that I will use one by one in my blogs as proposals to potential conferences and panelists. One area I have spent a good deal of time on is metering in data centers. I have visited several metering companies and interviewed their executives. For most of those companies, I have written a blog post:

• Arch Rock
• Modius
• OSIsoft
• Sentilla
  

In addition, I have talked to Liebert/Aperture and SynapSense. Although they are all in the same monitoring and metering space, each company has a slightly different angle. Some measure via their own sensors and aggregate the data for display. Others do not gather data by themselves but exploit the data from other sources. So the functions can be roughly classified into three categories: measuring (via sensor), aggregating, and analysis and display.

Also, in addition to measuring power consumption, different approaches can be taken to common functions like:

• alarm management
• asset management
• capacity analysis
• efficiency analysis
• air control automation

In the first panel, I would like to discuss the following so that a general audience can understand the needs of metering, learn which technologies are state-of-the-art, and assess the minimally necessary functions of metering. We would discuss the following:

• Why do data centers require monitoring and metering?
• What is a typical architecture for measuring, aggregation, analysis, and display?
• What are the minimally required functions of metering?
• What kind of standards should be defined for metering? What type of data should be collected? Granularity of data? Frequency of collection? Data formats (like XML)? Display formats?
• What extensions, such as e-waste and water, should be considered for existing metering?

An ideal panel will consist of three or four vendors and one researcher in the space. I plan to send this blog to the vendors I talked to before, at least one candidate conference, and a research organization, like a university.

Next, a proposal for a panel discussion of IT energy efficiency.

It began with a summary of the project from August 15, 2006, until April 28, 2010.

Its format has changed slightly over the three years but here are the most recent section heads:

• Release Status Update
• Deployment Status Update
• Schedule Update
• Deployment Update
• Budget Status—Expenditures by Workstream
• Budget Status—Benefits
• Issues/Risks Summary
• Program Metrics

After consumers complained about the smart meter installation in December 2009, the appendix section was added. Its contents change slightly from month to month, but the most recent appendix looks like this:

• Project Deployment Plan Progress
• Contingency Reconciliation
• Budget Narrative and Mapping by Workstream
• Meter Deployment and Customer Outreach Schedule
• SmartMeter™ Acronyms

This contains a lot of information, but I am not an expert. Here’s a rough summary of my findings:

• The key points are budgets and schedules. On their schedules, many factors are interrelated in a very complex manner. In spite of that, they seem to be put there as the schedule is planned.
• After consumers complained, consumer outreach programs were put in place, and some events were planned.
• The smart meter deployment is planned to be completed in the first quarter of 2012. The total number of smart meters is 10.1M.
• Some sections are blacked out. I wonder whether PG&E had something to hide when it made the report public.
• The metric section includes something like the customer complaint rate. It is around 0.029%, which is statistically negligible.
• Several service providers are mentioned in the report, including Oracle, IBM, EDS, PWC, and SSN (presumably Silver Spring Networks). BEA is mentioned in 2006; it was later acquired by Oracle.
• Some IT equipment is mentioned, such as Unix servers and mainframes. But neither Linux nor Windows machines are mentioned.

I can see why PG&E was reluctant to make this report public. If you are in the same business, you can exploit this information to compete with PG&E. The funny thing is that PG&E does not have to compete with anyone in their territory.

Finally, you can download the report from here.

The energy strategies are METI’s responsibility. Like any other country, including the U.S., Japan is keen on the development of solar technologies and their applications for generating power. Japan also is putting a lot of effort into enhancing battery technologies to smooth out the power generated by solar technologies, which are known to be unstable.

According to the Yomiuri Shinbun, one of the leading newspapers, METI published a target to increase battery life to 20 years (four times as long as the current version) and to cut manufacturing cost by 10% by the year 2030. METI will work with companies in this sector, and it plans to make the new technologies international standards.

So far in the renewable energy sector, the Japanese government plans to increase solar power generation to 28 M kW in 10 years. Well-known battery categories are for cell phones, PCs, and EVs. Batteries need to be smaller in size and larger in capacity. For residences, Sanyo Homes has begun selling a house with batteries. The battery cost is around $24K with the government subsidy. Moreover, Sekisui House and Osaka Gas are experimenting together on a house with batteries as well as solar and fuel-cell power generation.

The green conference was held at Tokyo Big Sight. It was big even by U.S. standards.

Tokyo Big Sight

I had a chance to meet with my blog readers, and it was nice to see them.

Pic of Mr. Inoue of NTT Facilities

Mr. Inoue’s presentation was about NTT Facilities’ data center technologies and practices. It consisted of a comprehensive set of technologies to save energy, such as aisle containment for better cooling efficiency and in-rack cooling. Later in the afternoon, many of the attendees (including yours truly) were bused to their research center and had a chance to see their research in progress. It was a very impressive presentation of their current research. More on their research will come, pending their approval.

Mr. Morishima (left) and Mr. Saka (right)

Some of my findings are the following:

• Where it stands: NEDO collected 324  (287 initially and 37 added) companies as a member of the Smart Community Alliance. This number shows the high interest in smart grid in Japan.
• The project will last for the four years from 2010 to 2013. This fiscal year is for preparation.
  
• The project is considered an experiment.
• Some of the details, like what to do with the infrastructure after the experiment, are undecided at this time.

• There are four working groups:

• International strategy: Encourages Japanese companies to enter the international market.
• International standardization: Works with the U.S. and Europe to set standards.
• Road-map setting: Charts a road map for technologies concerning smart grid. The U.S. NIST recently published its road map.
• Smart house: Unique to Japan.

In addition, I received details about the association, including its member roster. I offered to exchange information with them, and they are OK with it. I will probably meet with Mr. Morishima in the upcoming ConnectivityWeek in Santa Clara.

As I’ve reported in the past, EVs are hot in Japan, which is trying to be number one in the world of EVs. Technology-wise, it is too much of a big deal. The real problem is the lack of rare-earth metals. In manufacturing EVs, a lot of rare-earth metals are used, for example, in motors. I’d heard of rare-earth metals before without bothering to find what they are. 

According to Wikipedia,

Rare-earth metals are a collection of seventeen chemical elements in the periodic table, namely scandium, yttrium, and the fifteen lanthanides.

Rare-earth metals are produced mostly by China (more than 9%), which is catching up with Japan in the area of EVs. China’s exports of rare-earth metals have declined as it started working on EVs and needs of its own. That is making their price grow rapidly. Japan is now trying to locate rare-earth metals elsewhere, such as in Africa, but China is sourcing them there as well, competing with Japan.

If rare-earth metals are essential to EVs, what about the U.S.? I do not hear much about them in the U.S. China is spending a lot of money on renewable energies, and I wonder if the U.S. can compete with China for the next 10 to 20 years.

Electricity usage is the difference between the last reading on the meter and the current reading. Since PG&E does not work with Google PowerMeter or Microsoft Hohm, I log in to my account to see what’s going on. I guess in six weeks or so I should be able to see my electricity usage by the hour. It has been only three weeks since the meter was installed, and I can only see the information by the month. I guess it was like that before the smart meter.

I have covered the problem with smart meters in the past. Many consumers in central California complained about the meters’ accuracy when they received a bill two or three times as high as before the meter was installed. Some complained to the California Public Utilities Commission (CPUC) and others started a lawsuit against PG&E. CPUC took this problem seriously and hired an outside consulting firm to test the accuracy of the smart meters in question. PG&E also testified at the California Assembly and published the meter quality information.

Recently, CPUC took another step to satisfy consumers, according to the Central Valley Business Times.

The California Public Utilities Commission has ordered the state’s largest utility—Pacific Gas & Electric company, to make public all of its reports on implementation of so-called "smart meters.”

As a consumer, I am very much interested in this report. But as an analyst, I find it an invaluable source of information about how PG&E designed and architected the smart meter program. I can get hold of such a report since:

Under the new ruling, PG&E must provide a copy of its reports—or a link to a website where they are available—within two days to anyone making a request.

How soon does it happen? Actually, I found the link here.

• Presentation on U.S. data center problems and their solutions.
• Meeting with NEDO to understand what Japan is planning to do with the smart grid projects in New Mexico
• Meeting with CHAdeMO Association (rapid charging for EVs)
• Meeting with a publisher to discuss smart grid articles

I will report on each in next week’s blogs.

I copied a few pictures from there to this blog.

My monologue at the beginning. Photograph by Lisbeth Smestad.

With Dennis Symanski of EPRI. Photograph by Lisbeth Smestad.

All the pictures from the conference are here.

I moderated a discussion by five panelists:

• Isaac Kato, executive vice president and chief financial officer, Verne Global　(Iceland)
• Kimmo Pentikäinen, director of Environmental Information, Elisa Corp. (Finland)
• Sindre Kvalheim, chief executive officer, LocalHost (Norway)
• Eric Teetzel, manager, Green Energy Program, Google (U.S.A.)
• Dennis Symanski, senior project manager, EPRI (U.S.A.)

Isaac’s presentation was on Verne Global’s data center in Iceland, which has 100% of its power supplied by renewable sources (geothermal and hydro). In addition, the data center gets free cooling from the ambient air outside. Isaac lives in Boston and occasionally commutes to Iceland. He told me that the coldest days in Boston are colder than those in Iceland. It was a timely subject, since we feared the conference would be postponed after the volcanic eruption there only a few weeks ago.

Kimmo talked about power usage effectiveness (PUE), which is usually greater than 1.0 because it is defined as (IT power + overhead) / IT power. I use this representation after seeing Bill Weihl’s (Google) presentation.

If you generate power inside a data center compound, PUE may be less than 1.0. 

Elisa’s data center in Finland accomplished less than 1.0 PUE. I did not have time to discuss this with Kimmo at the conference, but I will follow up shortly.

Sindre discussed an underground minelike structure to host a data center, and it was quite interesting. Both Eric and Dennis talked about universally applicable DCEE practices.

Those of us who live in the U.S. tend to focus our attention solely on the U.S. market and trends, but we can learn a lot from other parts of the world. And DCEE is no exception. My (naïve) question of why they put so much emphasis on renewable energy was answered with "that is all we get.” If we did not have a conventional fuel source for data centers, what would happen? I am not sure it is a good idea to restrict data center power sources to renewable energy by some kind of regulation. However, with a reasonable dose of regulation, it certainly would encourage data center operators to find a place where they can access renewable energy more.

Two people had different opinions. One was Mark Thiele, who was director at VMware, but who is now involved in so much that I am not sure how to describe him. He told me that it might take a few years, but the surge of cloud computing would make CDC a practical solution. I was still a bit skeptical about it. I then talked to Alex Fielding of Power Assure, who showed me his research on CDC. He introduced me a few months ago to a person involved in NASA’s Nebula project. Between that person’s busy schedule and my trips, I did not succeed until recently in making an appointment with him.

I finally had a chance to visit NASA Ames, where long ago, I tried to take its research prototype to the commercial market. From 2000 to 2002, I was on the Ames campus (not in the secure area) with my partner. With a contractor badge, I could enter the secure zone as well. It was nice to be able to step back onto the campus. For those who do not know the security arrangement there, you need a sponsor (a NASA Ames employee) to enter the secure zone. Since it was early, I drove around the building where we used to have an office.

This is a picnic area just behind our building where my partner and I discussed how to deal with NASA

In any event, I met this person, who did not want to be identified or quoted but was willing to show me the Nebula box. Here’s a picture of it, which has been published many times and in many places in the past.

Nebula container

Since I promised not to identify or quote him, I am simply writing my comments here. By now it is not new to see a CDC. The container is a standard 40 feet, but the differences are in the contents. As Dean Nelson of Data Center Pulse told me, a CDC is just a big rack, and there are many ways you can design the inside. The servers can be mounted front-to-back or sideways. Cooling can come from the ceiling or from the floor. (Even though cooling was working inside the container, outside it was very windy and cold. Airside economizer could be kicked in, although I do not know if it had such a feature.) Unless you know these things, you cannot evaluate a CDC well.

We were standing outside the building where the Nebula box was placed. Around the building, there was a lot of space. Since it is a container, it can be lifted easily and placed on a truck for transportation. When you need more computing power, you can add containers with relative ease. When you no longer need a CDC, the container can be removed and taken to where computing power is required. So, in a sense, a CDC is greener than a data center housed in a real building.

The Nebula project’s objective is to provide a cloud computing environment for NASA and beyond. 

I poked around the Nebula website and found out many things. Some time ago, the U.S. government’s new CIO said something about cloud computing being superior to data centers. Several people criticized him for not being technically savvy, since cloud computing is housed in data centers. After reading several pages on the website, I think I now know what he really meant. Before, when one branch of the federal government needed to extend its computing power supply, it would build a new data center without sharing with other branches. A data center is expensive to build and maintain. It could run several hundreds of thousands of dollars to construct and run. With the advent of cloud computing, branches can share computing power.

Also very strange was the website’s emphasis on speed-to-implement. When we dealt with NASA eight years ago, we could not find "speed” and "NASA” in the same sentence. Obviously, a change in management did the trick. A small box alone could provide a large amount of computing power to NASA Ames and beyond. It is good to know our taxes are at work without being wasted.

As Andrew Fanara leaves the Energy Star program, EPA sent out the following letter to inform us of the new personnel in charge for several different fields of the program.

Hiroshi Kariatsumari of Nikkei Electronics elaborated on this news in January.

The Japanese smart grid project in New Mexico is led by the New Energy and Industrial Technology Development Organization (NEDO).  NEDO will work with the government of New Mexico, Los Alamos National Laboratory, and Sandia National Laboratories. The New Mexico state government plans to have five locations for the test projects. Among the five locations, NEDO will work on three projects in Los Alamos County and Albuquerque. A fourth project will summarize the results of the three test projects. NEDO has a budget of $30M (of which $18M is for fiscal year 2010). The projects will run from 2010 to 2013 (four years). Although it is not spelled out, I assume each participant invests its own funds for its own projects.

Here’s the summary from Kariatsumari’s article. The original article was written in Japanese.

1. Microgrid experiments in Los Alamos County

This project is an experimental energy management system (EMS) that balances traditional and solar-generated power in the distribution grid. The plan is to prepare 2MW of power generated by solar with 1MW of power storage. The EMS should be able to smooth out the distribution of the varying voltage of the generated solar power.

The following companies plan to participate:

• Toshiba will lead the project with EMS, real-time pricing, and communications.
• Hitachi will be in charge of installation, lead batteries, power conditioner, and transformers.
• Kyocera will be in charge of solar power generation.
• NGK Insulators will be in charge of NAS batteries.
• Itochu Techno-Solutions 
• NTT Facilities 
• NEC
• Fujitsu
• Mitsubishi Electric 
• JEF Engineering 

2. Smart house in Los Alamos County

This project prepares 3kW of solar power with 20kWh of storage, heat accumulators, and IT house appliances. In addition, smart meters, real-time pricing, and communications will be installed and tested.

The following companies plan to participate in this project:

• Kyocera, co-leader, will be in charge of the home energy management system (HEMS), solar batteries, power conditioner, batteries, and home communication.
• Sharp, co-leader, will be in charge of HEMS and home appliances.
• Itochu Techno-Solutions 
• NTT Docomo
• JEF Engineering
  
• NEC
• Panasonic

3. Microgrid in the commercial area of Albuquerque

This project installs batteries, cogen, fuel cells, and heat storage tanks, generates solar power, and verifies that the system works without accessing the outside power grid. The power is planned to be at 600kW. The power generation variation will be smoothed out with the batteries and EMS.

The following companies plan to participate in this project:

• Shimuzu http://, co-leader, will be responsible for EMS, the building energy management system (BEMS), solar power generation, batteries, and communications.
• Sharp, co-leader, will be responsible for EMS, BEMS, solar power generation, batteries, and communications.
• Meidensha, co-leader, will be responsible for EMS, BEMS, solar power generation, batteries, and communications.
• Fuji Electric will provide fuel cells, cogen, and gas engines.
• Tokyo Gas http:// will provide fuel cells, cogen, and gas engines.
• Mitsubishi Heavy Industries will provide fuel cells, cogen, and gas engines. 
• Furukawa Electric will provide lead batteries.
• Furukawa Battery will provide lead batteries.
• Itochu Techno-Solutions 
• JEF Engineering 
• Sumitomo Electric 
• Nissin Electric 

4. Summary of the test projects

This project will put together the findings from the three test projects in conjunction with the U.S. counterpart.

The following companies will participate:

1. Itochu and Accenture will prepare the overall summary,
2. Toshiba will prepare a summary of the projects.
3. Kyocera will analyze the PV distributed power sources.
4. Kandenko will analyze the preservation of distributed power sources.
5. Central Research Institute of Electric Power  will analyze cybersecurity and communications.
6. Toshiba will analyze the development of the model and simulation.

Naoki Inose (right) with Tokyo Governor Ishihara

He began by discussing the history of solar panel penetration into the Japanese market. The Japanese central government introduced a subsidy for the installation of rooftop solar panels in 1994, but it took some time before people started adopting solar panels. In the average household, a solar panel requires 3 kW of power. In 2002, the subsidy was $1,000 (roughly 100,000 yen) per kilowatt, so the average household received $3,000 (3 x $1,000). The whole cost of solar panel installation was then $30,000.

The following is the progress of adoption:

• In 2000, 20,000 households
• In 2001, 25,000 households
• In 2002, 38,000 households, with a subsidy of $1,000/kW
• In 2003, with a subsidy of $900/kW
• In 2004, with a subsidy of $450/kW
• In 2005, with a subsidy of $200/kW
• In 2006, 73,000 households, with no subsidy
• In 2007, 49,000 households, with no subsidy
• In 2009, with the subsidy reinstated at $700/kW

Inose compared Japan with Germany, which enacted a law to force utilities to buy the power generated by each household at a higher price than you normally buy electricity from the utilities. The utilities put an additional charge on top of the power bill to cancel out the difference. Because of this system and Japan’s abandonment of the government subsidy at the end of 2008, the total power generated by solar energy was 5.4 million kW in Germany but only 2 million kW in Japan. Japan, which once led the solar market, dropped from first place and was surpassed by Germany and the U.S.

Inose wrote that the TMG passed several laws to make Tokyo more sustainable, requiring a 25% reduction in CO₂ emissions by 2020, compared with 2000, and the generation of 1million kW of solar power by the year 2016. With more stringent control over CO₂ emissions, the TMG implemented a cap-and-trade system in April similar to the U.K.’s Carbon Reduction Commitment. The central government, pushed by this move, reinstated the subsidy for the installation of solar panels in 2009.

If you live in Tokyo and have an average house, you will receive $2,100 (from the central government, $700 x 3) + $3,000 (from the TMG) + $3,000 (from the ward in Tokyo). In total, the subsidy is $8,100. The installation cost is roughly $18,000. The net cost for you is around $10,000. Once this is done, you can sell your excess power: the power you generate minus what you consume at home. In Tokyo, 1kW is approximately $0.24, and the utility company buys 1 kW at $0.48. This comes to roughly $1,000 a year for the average house. So the ROI for the installation is about 10 years.

According to Inose, the reinstated subsidy and the introduction of excess-power selling are encouraging rapid growth in the solar panel market in Japan. Companie in the solar market are now ranked as follows:

1. U.S. (First Solar)
2. Japan (Sharp)
3. China (Sun tech)

Japanese companies gave up on bringing their ICT (information and communications technology) products to the U.S., but they feel they can compete in the U.S. cleantech field. Several large and medium-size Japanese companies will participate in smart grid experiments in New Mexico. I will report on that in a future blog.

However, the industry is trying to move to some standards, such as containers and modularity, to make the construction of a data center less expensive. A data center has so many components that it is hard to discuss the standardization of all of them. Let’s look at cold- and hot-aisle creation.

Rich Miller of Data Center Knowledge reported on the simple curtain method to create cold aisle by isolation. This method is very simple and inexpensive to implement and maintain. He referred to a recent presentation by Bill Weihl, Green Energy Czar at Google, in Green:Net 2010. (You can see his entire presentation here.)

Actually, I had occasion to tour one of Yahoo’s data centers, housed by Digital Realty Trust, in Santa Clara. See below. The picture of Google’s curtain implementation of cold aisle is in Rich’s blog. 

This is a very inexpensive way of creating cold aisle. Rich, in his blog, listed some vendors that sell such curtains. I am not sure that use of a curtain qualifies as a standard, but it is an easy way to implement airflow management. There are other ways of creating hot and cold aisle by having scalable containment like Wright Line’s. Some other data centers, like Sun (now Oracle), have their own containment for their racks.

It does not take a rocket scientist to think of an effective way to cool very hot servers and other IT equipment. You pinpoint your cooling to where the heat is rather than cooling the entire room. In Bill Weihl’s talk, he mentioned some important practices for energy efficiency, and one of them is airflow management.

Incidentally, my talk on U.S. data center energy efficiency in Tokyo on May 13 will be followed by a talk by NTT Facilities, an expert in constructing data centers with a solution similar to Google’s and Yahoo’s above. The cooling and its effectiveness requirements are universal all over the world.

Most attention is given to PG&E, whose smart meter quality data is given in this article. According to the explanation PG&E gave the California State Senate, only about 1% of their customers who received smart meters are having a problem. I am not sure "only about 1%” is a good way of describing the problem. From a business view, any customer may be just another account. But if this is you having the problem with your new smart meter, you would not be happy to be grouped into the 1%.

According to PG&E, the actual number of smart meters that had some type of problem is around 43,000. Of those meters, eight have a recording problem. The rest were improperly installed.

According to the California PUC, they received about 1,000 complaints on smart meters, most of which were installed in the Central Valley by PG&E. Lucky me. I am a customer of PG&E, but I am not in the Central Valley.

The PUC’s review work on problem smart meters will be completed in August. That might conclude this dispute. As a consumer, I certainly hope that my smart meter is not defective.

Let me recap it. The formula is as follows:

DPPE = (IT equipment utilization * IT equipment ability) /

(total power consumed – green power used)

Let’s look at the denominator. I warned in my earlier blog that if all the energy comes from renewable energy, then DPPE becomes infinite because the divisor is zero. It is well known that at this point not many data centers are fueled 100% by renewable energy.

Verne Global’s data center, which is located in Iceland, is one of the few that are so fueled. According to Isaac Kato, executive vice president and CFO, their data center is powered with hydro- and geothermal energy and does not use any dirty energy at all. DPPE breaks for their data center. DPPE for their data center is infinite. Even though I have not seen many data centers completely powered by clean energy alone, there are going to be such data centers emerging. DPPE needs modifications, and I encourage them to work in it.

The volcanic eruption in Iceland grounded planes in most European airspace for six days. The "Information is Beautiful” website reported interesting statistics.

They calculated the saving of CO₂ emissions resulting from flight cancellations during that period. The total is the saving realized by no flights minus the emissions caused by the eruption.

So this is the equation:

344,109 tons (saved) – 150,000 tons (added by the eruption) = 206,465 tons

This figure is only for European airlines. I know that many other flights between Europe and Asia and between Europe and the U.S. were cancelled. So the saving is much more.

This figure poses a few interesting questions. First, the cancellations caused many business activities and personal lives to be disturbed. But at the same time, we saved a lot of CO₂ emissions. We need to ask ourselves which is more important: to reduce CO₂ emissions or to preserve our business and personal activities. Of course, there should be some balance between the two. But how do you draw a line to decide the right balance? With the advent of the online world, we can communicate with people around the globe without leaving the comfort of our home. But that cannot replace physically being there with your business associates or loved ones.

Second, IT is equally to blame for the emission of CO₂. What if we decide to curb the use of IT in order to curb the emissions from its use? We need to assess how much IT is saving CO₂ emission by telecommuting, replacing business trips, and so on. In addition, IT saves power consumption (thus, less CO₂ emission) by optimizing supply chains and fleet management. Of course, IT consumes power as it functions. IT is used not only for absolutely necessary functions but also for entertainment and enjoyment like online video sharing and SNS. Again, some people may think the latter functions are absolutely necessary, while others can do without them altogether. Again, it is hard to draw a line between the two.

On top of this, if you throw in developed countries vs. developing ones, the discussion becomes much more difficult to deal with. I wonder whether we will eventually come to some kind of agreement?

So I gave him an overview and prepared a list of references to help him get up-to-speed. Understanding smart grid will probably also benefit my readers, even if they do not know anything about the power grid other than how to consume power.

First, I recommend a simple brochure from the Department of Energy (DOE). It describes the power grid, its problems, and the background from which smart grid emerged. If you have no idea what the power grid is and why smart grid is discussed these days, this is a must-read brochure (48 pages but easy to read).

1. Smart grid 101 by DOE

Be sure to check other information available from DOE.

2. Smart grid is the intersection of power, communications, and IT, and each and every existing standard from those areas must interoperate well. This means a tremendous effort of standardization. The National Institute of Standards and Technology (NIST) was given the task of putting the relevant standards together to work cohesively.

NIST’s smart grid page is here and here.  And the next site is also helpful.

NIST’s framework and road map are very informative.

NIST also runs an interoperability panel; more than 1,500 people are working on 22 stakeholder panels. They communicate via wiki.

 The 22 groups are listed here.
I belong to #14.

If you are curious about who is participating, this page lists everyone who belongs to the 22 stakeholder panels.

A few groups are given higher status because they are working on issues that need to be resolved as soon as possible:

SGIP Standing Committees

• Smart Grid Architecture Committee
• Smart Grid Testing and Certification Committee

SGIP Permanent Working Groups

• cyber security coordination task group (CyberSecurityCTG)

I am not sure about the difference between the standing committee and the permanent working group. Although this is not a complaint, the wiki site contains a lot of information and is, quite frankly, confusing.

If you are interested in security, NIST has published a draft cybersecurity report:

If you join one of the 22 stakeholder panels above, I have a few words of warning for you. It is relatively easy to become a member, but they do not keep you if you do not show any sign of participation. They conduct SGIP meetings (virtual and face-to-face), some of which are mandatory, and you are expected to participate in them. If you miss two consecutive meetings, you are suspended. You can regain your status by attending two consecutive meetings. I don’t know whether this requirement is too tough for you. The next face-to-face meeting for Northern California will be held in conjunction with Connectivity Week in Santa Clara.

Of course, the downside of a webinar is the logistics required to square it away. We tried to get movers and shakers in the data center space, and that means busy executives. Tracking them down and putting the agenda and presentations together was a challenge. The BrightTalk folks did a great job of taking care of the details. Yes, there were a few minor issues, but even they added some color to the session.

One of the key messages from this panel was the collaboration among a half-dozen data center organizations, such as The Green Grid, 7´24, SVLG, and Data Center Pulse. Since we could not invite everyone, we missed organizations like AFCOM and Critical Facilities Roundtable. I am sure I missed some organizations. We did not invite anyone from the government side, such as EPA and DOE.

The organizations we invited were created because of particular needs of data center market segments, such as users, vendors, and operators. I am not sure if any of them merge with other organizations. But the stakes are too high to keep data center power consumption running so high. Two heads are better than one. And certainly many more heads are working hard to solve the energy efficiency of data centers. I will keep covering this area.

I’ve talked before about how to select a location for a data center. The criteria include:

• Power
• Network access
• No natural disasters
• Climate
• Tax breaks
• Access to human resources
• Space
• Market

The first two are essential. Without power or network access, no data center can be constructed. It is most beneficial to discuss universally applicable disciplines for energy efficiency, but it is also interesting to know how four data center operators from Norway, Iceland, and Finland are exploiting their regional environments for energy efficiency. My April 28 panel will also have two experts from Google and the Electric Power Research Institute (EPRI), who will talk about universally applicable disciplines, with flavors of region-specific environments.

The panel consists of:

• Moderator: Zen Kishimoto, principal analyst, AltaTerra Research (U.S.A.)
• Isaac Kato, executive vice president and chief financial officer, Verne Global　(Iceland)
• Kimmo Pentikäinen, director of Environmental Information, Elisa Corp. (Finland)
• Sindre Kvalheim,chief executive officer, LocalHost (Norway)
• Eric Teetzel, manager, Green Energy Program, Google (U.S.A)
• Dennis Symanski, senior project manager, EPRI (U.S.A.)

Incidentally, Isaac Kato lives in Boston and sort of commutes to Iceland. Verne Global’s data center is completely fueled by renewable energies from hydro- and geothermal, and it exploits the cold air there. The data center is west of the volcano and has not been much affected by the eruptions. And other operators have a lot of interesting stories to tell about their energy efficiency efforts. This is a must-see event.

You can register here.

The conference by itself will be interesting enough, but they are making it even more exciting by having opening and closing receptions. The opening reception is scheduled for Monday, April 26, 6:00 p.m.–8:30 p.m., at Stanford University’s Wallenberg Hall, 450 Serra Mall, Building 160. I plan to attend this reception and would like to meet with some of my readers there. The Silicon Vikings and the Consulate General of Finland will host the networking event. Drinks and hors d’oeuvres will be served. Please RSVP and find more info here.

The closing reception is scheduled for Wednesday, April 28, 6:30 p.m.–9:00 p.m., at Dinah’s Garden Hotel and Trader Vic’s Restaurant, 4269 El Camino Real, Palo Alto, CA. Hans-Martin Friis Møller, director of Water, Energy, and Environment for Grontmij　Carl Bro, will give an informal presentation on the worldwide challenges of water and how new technology and programs are being implemented. The presentation will be hosted by the Danish American Chamber of Commerce, Invest in Denmark, and Innovation Center Denmark. Hors d’oeuvres will be served, and there will be a cash bar. Please register and learn more here.

While I was in Japan recently, the news of the collaboration between the GridWise Alliance in the U.S. and Japanese organizations was reported. h

The following is GridWise’s mission:

To facilitate the effective collaboration among all stakeholders and to promote, educate, and advocate for the adoption of innovative smart grid solutions that will achieve economic and environmental benefits for customers, communities, and shareholders.

According to Transmission & Distribution World, the usual suspects on the Japanese side signed the MOU with GridWise Alliance. They include:

• New Energy and Industrial Technology Development Organization (NEDO)
  
• Smart Community Alliance (no specific website, as it is built by NEDO)

NEDO has been involved, along with several Japanese companies, in smart grid construction for a few communities in New Mexico.

While researching a way to connect my smart meter with PowerMeter and Hohm, I found this article by TreeHugger. It says that PG&E wants to stay open, meaning that they will not connect their smart meters with either Google PowerMeter or Microsoft Hohm. So I looked at the Google PowerMeter site for their partners. 

Sure enough, PG&E is not listed. Actually, PG&E is waiting for the Open Smart Grid group to decide on a standard interface. OK, then to PG&E’s smart meter site. I need to sign up with them to get my power usage information online.

For that, I need to consult with my VP of Operations at home. Stay tuned for what we decide to do.

The sizes of the data centers he discusses range from 400K sq, ft. to 1M sq. ft. Rich and his team’s criterion for the top ten is that a data center must have its own building and not be part of a larger building. Also, their More Worthy Contenders addendum is worth going through, as they discuss data centers for Google and Apple.

The following is his top ten (#1 being the largest):

• 10. The SuperNAP, Las Vegas (Switch Communications)
• 9A and 9B. MicrosoftData Centers in Quincy Washington and San Antonio
• 8.CH1, Elk Grove Village, Ill. (DuPont Fabros)
• 7.Phoenix ONE, Phoenix (i/o Data Centers)
• 6. MicrosoftDublin (Microsoft)
• 5.Container Data Center, Chicago (Microsoft)
• 4. NGD Europe, Newport Wales (Next Generation Data)
• 3.The NAP of the Americas, Miami (Terremark)
• 2.Metro Technology Center, Atlanta (Quality Technology)
• 1. 350 East Cermak / Lakeside Technology Center(Digital Realty)

I’ve participated in a tutorial on SuperNap and felt it was pretty big, but it turned out to be #10 on the list. We have four Microsoft data centers, as they publish on their data centers while Google does not publish much about theirs. (See "More Worthy Contenders” for details.) The list contains mostly U.S. data centers or data centers owned by U.S. companies. A comment on this blog stated that @Tokyo in Japan has about 1.5M sq. ft. of space and is located in the district of Toyosu.@Tokyo was built by a division of Tokyo Electric Power Company. 

The Toyosu district of Tokyo is growing rapidly, lacking enough primary schools because of people relocating from elsewhere. It looks very modern, like an American city. The Tokyo Metropolitan government was planning to move the Tsukiji Fish market to this area. 

So many sightseers have been interrupting business in the market that all visitors have been banned for a month.

My search found the New Energy Foundation (NEF) working with New Energy and Industrial Technology Development Organization (NEDO). I met Mr. Makoto Okuda, the director of the Fuel Cell Department of the Research and Planning Center. NEDO has eleven projects under way, including development and validation of solid oxide fuel cell (SOFC) and polymer electrolyte fuel cell (PEFC) technology, and hydrogen technology and infrastructure development.

Mr. Makoto Okuda

The PEFC type is for small-scale systems, and the SOFC is for large systems. The PEFC is planned for automobiles and small power systems for residential use. Since I want to compare Bloom Energy with what’s happening in Japan, I will focus on SOFC in this blog.

NEF classifies SOFC in these four categories:

• Residential—between 1 kilowatt and several kilowatts
• Commercial—between several kilowatts and several hundreds of kilowatts
• Industrial—between several hundreds of kilowatts and several megawatts
• Large installation—several megawatts and more

Small-scale SOFC will accomplish $10,000 per kilowatt in 2012 or so; a comparable solution is about $700,000 per kilowatt. A good-size data center would require an industrial-size fuel cell, which should be available around 2012.

Right now, NEF is conducting validation of the residential version of the SOFC technology. Participants include large gas companies like Osaka Gas and Tokyo Gas, electric power companies like Tokyo Electric Power Co., Nippon Oil Corp., and TOTO. Except for Nippon Oil, which uses LPG and kerosene, the fuel is natural gas. Fuel cell system vendors include Kyocera and TOTO. 

I discussed the following with Mr. Okuda:

• Amount of investment
• Speed to market
• Elaborate validation

The Japanese government is investing $12M in SOFC technology; in contrast, Silicon Valley–based Bloom Energy has raised $400M in funding so far. Also, the speed-to-market is significantly different. Bloom Energy, in competition with many others, needs to reach the market at utmost speed, while in Japan several companies need to work together under the guidance of the government.

Finally, the Japanese effort spends lots of time on validation. The SOFC technology runs its system under high temperature (700–1,000 °C), and it is necessary to make sure the system does not break after operating at high temperature for a long time. Mr. Okuda did not question the quality of Bloom’s fuel cell, but he was interested in reviewing the results.

Finally, I may be able to visit Bloom Energy soon and will ask for such data.

Power crisis and energy efficiency, as they relate to data centers, have been discussed at many conferences and meetings. In this panel, we will take a different angle on three topics. One is IT energy efficiency improvement. A 2007 EPA report looked at five different scenarios for the past, present, and future of data center power consumption. Each scenario is defined by the energy efficiency of both IT and facilities. Of course, the facilities side is important, but the very reason to have IT equipment is to support business needs. IT comes before facilities, and we need to consider IT equipment energy efficiency as such. Each of the five scenarios has a different degree of the following to achieve a different level of energy efficiency:

• Turning off of unused servers

• Virtualization

• Power management

• Hardware refresh

• Storage equipment efficiency

The EPA study was conducted in 2006. In the four years since then, IT energy efficiency technologies and practices should have become more innovative. First, we would like to explore that. Second, several data center organizations advocate different metrics, technologies, and practices for energy efficiency. Doesn’t it make sense to have one set of standards for everyone to use, eliminating the confusion? Third, to produce a single set of standards, all these data center organizations must get together and collaborate rather than continuing individually to develop their own, all of which may be slightly different.

Summing up, these are the three topics:

• IT energy efficiency optimization

• Standards

• Collaboration

We are fortunate to have four experts in data centers and data center energy efficiency for this panel:

• Richard Garrison, board member of the 7´24 Exchange California chapter and senior principal at Alfa Tech Cambridge Group

• Jon Haas, director of Eco-Technologies at Intel and board member of The Green Grid

• KC Mares, co-chair of the SVLG Data Center Energy Efficiency Demonstration Program and president of MegaWatt Consulting

• Dean Nelson, founder and chairman of Data Center Pulse and senior director of eBay

We expect a lively conversation on these topics. Again, you can join us for this webinar at your desk at 11 a.m. PDT on April 21. See you then.

Am I taking the train mainly because of it greenness? To me, the main reason is that it is convenient. Even if the flight takes only 45 minutes, I need to be at the Haneda airport (which is more convenient than the Narita airport) at least an hour prior to departure time. In the case of the bullet train, I can get to the Tokyo station and purchase a ticket as late as five minutes before departure time. The bullet train runs every ten minutes or so, and if I miss one, I can easily catch the next one without waiting long. Another reason is that I hate flying, especially after flying twelve hours from the U.S.

Even though trains are more energy efficient than airplanes, they consume a lot of power. JR East owns its own power plants, in addition to sourcing power from Tokyo Electric Power Co. During the morning and evening rush hours (when commuters flock together to use trains), JR East uses hydroelectric power to supplement their power needs. During off-peak times, they pump water to a higher elevation and release it to generate power when it’s needed.

The train service is great for a country of Japan’s size. The high-speed service being considered between San Francisco and Los Angeles is a good idea. But how do we get to the station to board the train? And how do we get to our ultimate destination once we get to the destination station? I wonder.

GIPC’s definition of DPPE is somewhat convoluted, defining submetrics first:

• IT equipment utilization (ITEU)
• IT equipment energy efficiency (ITEE)
• PUE
• Green energy coefficient (GEC)

Conceptually,

DPPE = (IT equipment utilization * total IT equipment ability) / (total DC power consumption – green energy used)

= IT equipment real use / total power consumption

More precisely,

DPPE = ITEU * ITEE * 1/PUE * 1/ (1-GEC)

= (IT equipment utilization * IT equipment ability)

/ (total power consumed – green power used)

So they are saying that DPPE is a ratio of the actual IT equipment’s ability (because some of its total ability may not be used all the time) against the dirty power consumed.

GIPC claims that they came up with this metric after interviewing many data center operators and related companies.

Let’s look at two other metrics considered as improvements to PUE: data center energy productivity (DCeP) and corporate average data center efficiency (CADE). First, DCeP, under consideration at The Green Grid, is similar (but not identical) in spirit to DPPE.

DCeP is defined as:

useful work produced / total power consumption

This is different from DPPE in two ways:

1. DCeP does not take the type of power (dirty or clean) into consideration.
2. DCeP considers the output that is really useful.

Although I understand that the consideration of clean vs. dirty power may make sense, in reality we are not even close to the use of clean power for data centers yet. Although it is not possible to have all the power supplied by clean (green) power, if a data center is fueled by clean power alone, this ratio becomes infinity. Any metric that has a potential to become infinity may not be appropriate.

As for the second point, DCeP takes the software aspect somewhat into consideration. DPPE seems to simply count MIPS for given computing, regardless of what is produced. In an extreme case, you can run meaningless software to keep a server occupied and produce useless results. But I think DPPE is using this as a proxy for the useful work, as in DCeP. In addition, defining useful work has been a problem. That is why several proxies have been proposed for DCeP.

CADE, on the other hand, was developed by McKinsey, in conjunction with Uptime Institute.

CADE is defined as:

CADE = facility efficiency (FE) * asset efficiency (AE) where

• FE = facility energy efficiency * facility utilization
• AE = IT energy efficiency * IT utilization

This also is somewhat similar, although not identical, to DPPE.

Like any metric, DPPE has its merits and demerits. The Green Grid and other organizations have their own metrics for improving PUE. If DPPE is promoted as the world standard for energy efficiency at data centers, GIPC needs to have open discussions with The Green Grid and others. Everyone has an opinion on energy efficiency metrics, and no single metric can express energy efficiency perfectly.

In my opinion, Japan has been following many of the U.S.-led initiatives and technologies to date. Whether or not DPPE is accepted as a world standard for data center energy efficiency, it is a good thing for Japan to express its view outside of Japan.

1. Background information for smart grid
2. Overview of smart grid
3. Technologies used in smart grid
4. Smart grid progress and status
5. Problems and challenges
6. Future
7. Small and medium companies’ business opportunities
8. 
   
9. Some of the attendees were people from:

• Canon
• Fujitsu
• Hitachi
• Investment companies
• Mitsubishi Electric
• NEC
• NHK
• Nikkei
• NTT

Hopefully, my talk gave a good tutorial on smart grid in Japan.

So what is the point of this blog? Aside from the usual mumblings, it is to point out that IT can be used extensively to avoid business trips, reducing GHG emissions. I physically moved myself by flying on an airplane. I had a minor family matter to take care of that could not be done virtually online. However, I can conduct business as if I am in California. Yes, IT needs to be more energy efficient, but IT is making other things greener, for sure. I am sure there must be a study of the balance between IT’s responsibility in terms of GHG emissions and the savings accomplished by IT. Is there such a quantitative study? Let me know if you know of one.

This includes the U.S., Europe, and Japan. The agreement is by The Green Grid, the U.S. Department of Energy’s Save Now and Federal Energy Management Programs, the U.S. Environmental Protection Agency’s Energy Star Program, the European Union Code of Conduct, Japan’s Green IT Promotion Council, and the Japanese Ministry of Economy, Trade, and Industry’s Green IT Initiative.

It is very strange that I cannot find much news of this in the Japanese media, except for Nikkei’s web version. Even the Nikkei’s report on this is sketchy.

I will ask my contacts at JDCC, a consortium of data center operators in Japan, about this. It is very hard to believe. Even though PUE is not the perfect metric, it is a step in the right direction to apply the same energy efficient metric to data centers. More to come on Japanese data center operators’ take on this.

Even though this program is not widely popular, it is interesting to see a TV program on cloud computing for mass consumption. The interest in cloud is very high here in Japan, but so far there has been no real implementation of cloud computing. The Japanese government has a plan, known as Kasumigaseki Cloud, to implement cloud computing for its administration. I have not heard its real use yet. The private sector story is very similar. At the end of last year, Amazon Web Services landed in Japan and started providing its cloud platform.

When a Japanese company wants to use cloud services, it tends to go to one in the U.S. Prof. Murai’s TV program gave an example of this. Fourteen people in an architect’s office are using cloud services across the Pacific Ocean. Their applications are mostly for data archiving and sharing without mission-critical, real-time applications.

When I access web servers and other services in the U.S. from Japan, unless the application is very delay sensitive, it is quite usable. The wider bandwidth of broadband here may have an impact. The bandwidth is probably 10 times or more in Japan than it is for a typical service in the U.S.

Up to this point, the story is great. Since this is a blog on energy efficiency, I need to pose a question like this. If Japan is using cloud services in the U.S. because of their availability, usability (fast enough to access), and cost (Japanese equivalents of any U.S. IT equipment and services are usually more expensive), what about the traffic across the Pacific Ocean (apart from security and national interests)? Remember Brian Lillie’s keynote address at the recent AFCOM Data Center World. Equinix’s CIO had an interesting slide (see below).

Internet traffic grew tremendously from 1997 to 2007, as the slide shows. If many countries from Asia and Europe utilize cloud services in the U.S., the traffic will increase further. This is probably good for the U.S. economy, but is it green? In order to sustain such an amount of traffic, the cables placed at the bottom of the Pacific Ocean need to be strengthened and expanded. That will take ships and other resources. That will probably put some pollutants into the ocean. On the other hand, this may cut business trips over the Pacific and Atlantic oceans. It is very hard to say whether IT is helping the Earth or polluting the Earth. It is necessary to have a balanced discussion on the merits of IT and cloud computing rather than an emotional discussion without logic.

I did not plan to remember my young days in this blog. In a recent conference on sustainability, someone said that Japan is far ahead of the U.S. in terms of energy conservation. She was referring to an extension power cord with switches. It may not make sense when you hear about it, so here’s what I mean.

Each outlet in the extension cord comes with a switch. In this way, you can control each outlet. In this picture, the leftmost and rightmost outlets are on, while the middle two outlets are disabled. Those who are really opinionated may say that electricity is being wasted on the indicator lights. Panasonic (the vendor of this extension cord) claims that each light consumes only 0.16 W. This extension cord is good up to 1,500 W. According to the specification, it is good for 125 V with 15 A (1,875 W).

When I first saw this design (it was introduced only ten or so years ago), it did not capture my attention the way it did when the person at the sustainability conference mentioned it. In Japan, some people say Japan is behind the U.S. and Europe because Japan does not have this or that. It seems that the same is true in the U.S., where some people say that Japan and other countries are ahead of the U.S. in terms of this and that. That really makes me smile. Knowing both the U.S. and Japan, it is hard for me to say which is better. It is not that either the U.S. or Japan is better but that the two countries are very different. Coming from an energy-scarce country like Japan, I can see that the U.S. still enjoys the energy-abundant life style.

By the way, if this type of the extension cord is so new to the U.S., I am bringing it back to show it to everyone to gauge their reaction. One more thing to look forward to!

With this announcement, Mitsubishi, known to have its own EV, the i MiEV, cut its price down to $32,000 with the help of the government’s stimulus fund. So in Japan, the battle has moved to the below-3-million-yen range, which significantly loosens up your wallet.

GM, which was planning to market its own EV, the Chevy Volt, has not announced its sticker price but will need to reprice it after this announcement.

In an informal chat, a utility expert told me that EVs will have difficulties spreading in the U.S. the way they are in Japan and Europe, especially in the western parts of the U.S. Of course, a town like Berkeley is an exception. He was of the opinion that EVs are suitable for smaller and more populous areas, like cities in Japan and Europe. In the U.S., specifically in the west, cities are spread out, and driving distances are longer than they are in the eastern parts of the U.S.

Both EVs from Nissan and Mitsubishi must recharge after running 100 miles. In California, some people have longer commutes than that. So you need charging stations to recharge the batteries. You do not want to spend hours charging your batteries. so rapid charging is a must for the widespread use of EVs. In Japan, CHAdeMO, (http://www.chademo.com/indexa.html) an organization to promote the proliferation of rapid charging, has been set up. The main players in this organization are Toyota, Nissan, Mitsubishi, Fuji Heavy Industries (maker of Subaru), and TEPCO (Tokyo Electric Power Company). It is setting up a standard, the CHAdeMO protocol, for rapid charging. I may visit them if I have time after visiting the SOFC cell promotion organization.

The Tokyo version is very similar to the CRC, and both start on the same day.

When Tokyo’s environmental protection law was announced, rumors flew that many data center operators would leave Tokyo and take refuge in the neighboring prefecture (state) of Kanagawa (where Yokohama is). It is still early, but I have not heard that any data centers are moving as a result of this law. I will find out what’s going on when I meet with the Japan Data Center Council later. 

PUE has been suggested as a way to get a good feel for data center energy efficiency. It is defined as:

(total power consumed at a data center) / (total power consumed by IT equipment)

PUE has been around for several years, but it has taken time to sink in to the data center market. Every time I go to a data center conference, an informal survey is taken to see whether data center operators know PUE and use it for their data center operations. It seems to be catching on these days. Digital Realty Trust’s recent survey of 300 enterprise data center operators (those with at least $1B in revenue or more than 5,000 employees) revealed that less than 2% did not know what PUE was.

It is well known that each data center is unique and that sometimes it is hard to apply the same practices in different ones. Having said that, it is also very important to have a common language, such as energy efficiency, for talking about the variety of data centers. In a recent panel, Jon Hass, technical committee chair of The Green Grid, emphasized the importance of collaboration and a common language for the data center industry. I wholeheartedly agree with Jon on this point. I will moderate a webinar on the topic of data center energy efficiency on April 21, and Jon will be one of the panelists. We would like to extend the discussion of EPA to cover standards and collaboration in general in this webinar. There are many players, organizations, and governmental entities working independently on the data center energy efficiency issue. But it is time to put our heads together to solve the industry and the nation’s problem to tackle the power shortage.

Finally, although PUE is not perfect, The Green Grid is working on formalizing a way to measure it. The Green Grid is also working closely with EPA, which plans to publish an Energy Star rating for data centers in June. As Jon said in the recent panel, PUE will probably be replaced by IT effectiveness per watt. IT means both hardware and software, although people tend to consider IT effectiveness in terms of hardware boxes. See my previous blogs on this issue.

Mark Thiele, an expert in data center/cloud computing, mentioned this company. He was nice enough to introduce me to Dave Roberts, VP of marketing at ServiceMesh. Incidentally, I knew Dave a little bit when he was at a networking company in Silicon Valley. He moved from Silicon Valley to Austin, Texas, where most of ServiceMesh’s development team members reside. The meeting was conducted using a Web meeting service.

Dave Roberts

It took me a while before I could understand ServiceMesh’s business. They are a software company and provide professional services as well. In short, ServiceMesh helps large enterprise customers transition to an on-demand, everything as a service IT operating model. Their software product, Agility Platform, provides an end-to-end governance, workflow, policy management, and automation system for deploying applications in clouds. Their software reads in all the application information with their IT equipment information (CMDB if available), SLA, and security requirements, and spits out a report. The report contains feasibility information as to whether each application should be run in a cloud. If particular software requires very stringent response time, for example, it may not be suitable in a cloud environment. As shown in the following figure, ServiceMesh’s software can interface with a variety of different cloud service providers, including internal service providers building private clouds. ServiceMesh makes it easy for enterprises to use the right service provider for each application, subject to the specific SLA, security, and accountability requirements. Once the correct application stacks have been identified and built, ServiceMesh automates the deployment process, using cloud-provider APIs to start, stop, and manage application virtual machines in the target cloud.

The name of the software product is Agility Platform, and it consists of four major components:

• Agility Planner generates a report as to which applications are suitable for cloud computing and provides the basic configuration information to Factory.
• Agility Factory takes what was produced by Planner and allows a user to create a set of stacks suitable for each enterprise application. New stacks can be created from existing stacks already stored in CenterPoint.
• Agility CenterPoint stores stacks that have been verified and used before in that enterprise. It serves as a repository and collaboration point for stacks.
• Agility Manager deploys stacks into one or more cloud providers (including private clouds) and monitors them over time.

This is shown in the next picture.

In addition, ServiceMesh provides identity management, security by VPN, firewall, host intrusion detection, and antivirus.

Even though cloud computing is maturing, enterprises are still slow to adopt it. Major reasons given for that are the lack of security and SLA, vendor lock-in, and so on. I think ServiceMesh’s offerings are unique and interesting for a few reasons.

1. They work with leading cloud computing providers to improve their services to suit an enterprise’s stringent requirements. This alone could be a major contribution.
2. Cloud computing is totally new to most enterprises, and they do not know how to start with it. ServiceMesh provides them strategic advice on how to use it. The first thing to do is to inventory IT equipment and applications. Not all applications are suitable for a cloud environment, and this information alone is very useful to know.
3. The Agility tool suite provides end-to-end governance and policy controls around the deployment of applications into the cloud. This helps IT keep control of key application and data assets and ensure that they comply with all legal, regulatory, and business rules while still accessing the benefits available with cloud computing.

To target enterprises that are new to cloud computing, it is necessary to satisfy their needs and provide automation for ease of conversion. For the first point, ServiceMesh teams up with leading cloud providers to solve those concerns. For the second point, ServiceMesh’s automated tools help and accelerate the adoption. They seem to be signing up large enterprise customers rapidly. Of course, only time will tell how their business models work out.

This is a smart grid 101 seminar. After I introduce what is happening with smart grid in the U.S., I will discuss a business opportunity for small to medium-size ICT companies. When I gave a similar seminar in January in Tokyo, the place was packed, and some people could not register at all. As interest in ecology and green business is soaring, smart grid is getting a lot of attention. EVs and solar panels are at the center of smart grid in Japan.

Fuel cells can be used for EVs as well as for energy storage in the ecosystem of smart grid. While I am in Japan, I will try to see one or two fuel cell consortia to learn more about what is happening with fuel cells. Tokyo Gas has a service called EneFarm that is a fuel cell–based service for home. I plan to visit them as well.

My next blog post will be from Japan. In the age of the Internet, geography does not matter. I have as easy access to the Internet in Japan as in the U.S.

I’ve covered this topic more than once after attending Mark Bramfitt’s presentations in the past. The best scenario for utilities is that they maintain their generation capacity without expanding (expensive to do) or buying fuels on the spot market (expensive) to cope with the peak demand. When I first heard about this, it puzzled me a lot. It is because utility companies sell power to make a profit. In a normal business, the more you sell your product, the more profits you receive. In the energy business, it does not work that way. In addition to avoiding expansion and purchasing extra fuel at a high price, utilities are under a lot of pressure to go green. In several states, like California and Connecticut, the profit and the amount sold are decoupled.

So in short, utilities would like to pay you money to curb your power consumption at your data centers. This session was given by A.J. Howard of Energy Market Innovations. First he distinguished rebates from incentives. Energy rebates are given after the energy improvement is done, while incentives are paid before the improvement.

Also, it is important to keep in mind that the utilities payment is for the incremental cost of more efficient technologies or designs. Usually, there are two programs: energy efficiency and demand reduction.

As for a data center, many of its components are targets for the rebates:

Probably the most famous one is the one for server virtualization.

I am omitting the details, but utilities all over the U.S. are participating in energy efficiency rebate and incentive programs. A.J. showed a map of utilities that participated in those programs (see below). This is not an exhaustive list of all the utilities. If you want to find out more, contact your local utilities. They may belong to the Utility IT Energy Efficiency Coalition, which is committed to developing programs for IT and data centers.

PUE = (Total power delivered to data center) / (Total power consumed by IT equipment)

the denominator is always smaller than the numerator.

However, if you read James Hamilton’s blog, we cannot laugh at the marketers anymore, although I do not think they ever knew what James is talking about. When James blogged about PUE of less than 1.0 before, I asked him to elaborate on it. He told me that he would do so sometime in the future. He explained it in this blog:

I did not notice that post until now.

In short, energy delivered to the facilities is converted into heat generated by IT and facilities equipment. According to Neil Rasmussen of APC, almost all (99.99%+) of the power that enters a data center is converted into heat. This is very well illustrated in Figure 2 on page 4 of the PDF.

In the average data center, the heat generated by operating IT equipment (such as servers) is ejected into the air without recycling the energy represented by heat. James asks, What if this heat is used to grow crops to create biomass renewable energy? Such energy could be reused to fuel the data center.

He gave an example and did a back-of-the-envelope computation to come up with a PUE of 0.9. A data center with 1.2 PUE wastes 16.67% of its energy in power transmission and cooling. To make the computation easier, let’s assume 120 units of power delivered to a data center and 100 units consumed by IT. The 16.67% comes from the loss (20 units) divided by the total (120 units). At least 100 units (83% of the delivered power) generates heat inside the data center, where it is easier to capture. Almost all of the 100 units are converted to heat. If we can capture this heat and convert it to renewable energy via biomass, and let us assume that our energy conversion rate is 30% (30 units)，then the PUE becomes 120 / (100 + 30). This yields 0.92, which is less than 1.0.

I’ve conducted several interviews with companies in the data center energy efficiency space. Sometimes, repeatedly bumping into someone from one of those companies at industry conferences and meetings prompted my interview. Sometimes, I have been introduced by someone who knows someone at such a company. My encounter with Craig Compiano of Modius was very different from the others.

As I have been reporting in my blog, I attended Data Center World in Nashville, TN, in early March. Craig gave a talk there regarding the importance of measuring and metering for data centers, from the Modius perspective.

At that time I was completely confused about this space (metering/measuring). Well, you put your own sensors or someone else’s at strategic locations (including both IT and facilities equipment) in a data center. Then, you collect and gather data, then you aggregate and analyze them. Finally, you display them in a dashboard. I never belittle the efforts to make a product out of any technology, but this seemed pretty straightforward—maybe too straightforward to consider as an independent space.

So after Craig’s talk, I went to him and asked this question. Maybe my tone of voice was not very friendly, but he gave me his answer. If I remember correctly, he told me that I needed to evaluate each company’s offering from many angles, including comprehensive device coverage, supportability, and scalability. Obviously, I was not convinced, and my face showed it. Craig said he could spare more time if I wanted. I thought I was going to do that if I had time. Then, at the Modius booth on the show floor, I bumped into Adam Waitkunas, whom I got to know by attending a few data center conferences and meetings. I told Adam my problem of not being able to differentiate those metering companies. He told me that he could arrange a meeting with Modius. So that was why I got to talk to Craig again on a hot San Francisco day.

When we sat down in his conference room, I was eager to find out the differentiation. He started drawing pictures on a whiteboard, as shown below.

Craig Compiano

It was deja vu. I remembered this scene from somewhere. Then it came to me. More than fifteen years ago, Jim Clark gave me the same kind of chalk talk about Mosaic Communications (later Netscape). Because of this deja vu, I knew that Craig is a technologist as well as a chief executive. And he certainly is. His view was quite interesting. The following is my version of his picture in a more abstract (less artistic) way.

Modius view of metering/measuring space. The box in light beige has been implemented by Modius, the one in light green is someone else’s, and the box in light red is being implemented.

His explanation follows.

What Modius did first was to build a middleware infrastructure to support several functions required at data centers. The middleware is enclosed by a dashed line in the picture. This middleware infrastructure is the cornerstone of the Modius technology; it allows scalability to support any number of data center locations around the globe. The database is also an important piece of the equation; it needs to collect, organize, and retain a huge amount of data for analysis, display and sharing with various applications.

A set of necessary functions (or applications) have been implemented on top of the middleware infrastructure. The five leftmost functions are those typically implemented at most data centers today:

• Real-time Alarm Management: Data set by various sensors regarding health of equipment and power consumption
• Asset management: Inventory of IT and facilities equipment
• Service desk: Help desk function
• Capacity analysis: Current and future power and cooling capacity
• Efficiency analysis: Real time analysis of the efficiency of DC power and cooling assets.

The last box is automatic air control, which controls the airflow and temperature of CRAC units on the basis of the inlet air temperature at each server in the data center. I am familiar with this technology because I edited the case study for the California Franchise Tax Board data center. Federspiel is marketing this technology. See in here under the header of "Control of Computer Room Air Handlers Using Wireless Sensors - Franchise Tax Board"

Modius is currently working on this technology, and has a product plan to extend their functions even further, but Craig is not sharing more information on that at this time.

Back to my basic questions. What is the differentiation? With this picture on the board, it was straightforward.

• Enough functions on top of the middleware structure. This is a necessary condition but is it sufficient? I am sure new applications will be added over time all fed by the same real-time source data. How many more functions are required to be enough? This needs more thinking and analysis.
• The infrastructure, including the middleware. Can it scale as an enterprise wants to add more locations and more applications into a single converged system? Does it have an easy interface to add new devices and equipment as they become available in a low-cost manner?

What about the big four in network/system management? Are they coming into this space anytime soon? Craig does not think so. The complexity and disparity of infrastructure assets being addresssed by metering/measuring companies and the highly varied consumers of the metric data is not necessarily tightly linked with network/system management. But who knows? If and when this space matures, this space may be merged into network/system management to form a new space.

So far Craig’s explanation was very useful to analyze the market space. If I had not approached him after his talk, would he have taken the time to share his thoughts with me? The talk with him has solved some of my confusions.

The meeting consisted of four parts: fuel cell 101, Fujitsu’s actual use of fuel cells, panel discussions, and a tour of Data Domain’s data center.

Jesse Smith of Nova Partners gave a pretty precise presentation on what fuel cell technology is all about.

Jesse Smith

I cannot reproduce his whole discussion here, but he talked about three major fuel cell vendors. I hear there are hundreds of vendors in this space, but he picked only major ones that use different technologies.

Jesse talked about the California Public Utilities Commission's Self-Generation Incentive Program. Without going into details, if you generate 200 kW, your rebate from CPUC is in the range of $500,000 to $1 million. He also said that the rule of thumb for deciding if this makes sense for you is as follows. Get a price (A) for 1 kW and figure out how many BTUs are required for generating 1 kW of electricity. Get a price (B) for that BTU. If the difference between A and B (A being larger) is three cents or more, then it is good to go with fuel cell energy.

Following Jesse’s talk, Ted Viviani of Fujitsu discussed his experience in implementing fuel cell technology for their campus in Sunnyvale, CA, where I have visited several times in the past. His talk was full of data, and I hate not to be able to show some of the details. (I might discuss those details in a future blog.) The following slide compares his fuel cell solution with Google’s PV solution.

In this slide, he shows his box and the energy comparison:

Google’s PV required 400,000 sq. ft. to produce 2,329,380 kWh per year. Ted’s system used 164.8 sq. ft. and produced 1,654,608 kWh. In 0.04% of Google’s space, his system produced 71.03% of Google’s energy. It is certainly very energy effective. He also showed a map of United Technology’s deployment (below). Japan is full of dots, considering that it is almost one-thirtieth the size of the U.S.

The panel session consisted of the following speakers:

Moderator:

• Jesse Smith, Senior Project Manager, Nova Partners
  

Panelists:

• Subodh Bapat, Vice President and Distinguished Engineer for Oracle
• Ted Viviani, Building Operations Manager, Fujitsu Management Services of America
• Michael Bangs, Director of Facilities for Adobe

From left: Jesse Smith (standing), Subodh Bapat, Ted Viviani, and Michael Bangs

Again, only some of the highlights are discussed here.

• The boot time of a fuel cell system is from 8 to 18 hours. Thus, it cannot be used in place of a generator.
• The fuel cell system should be used as a baseline system and supported by another power source, such as a utility.