Posted by Michael Bullock on Tue, May 18, 2010
AFCOM recently suveyed 436 member data center sites on eight of the hottest topics ranging from Greening and Data Center Consolidation to Performance Monitoring and Cyber Terrorism.
-
- 71% are actively engaged in greening
- Power Recovery - 60%
- Cooling Recovery - 51%
- Expansion
- 60.3% require additional data center capacity within five years
Please view the tabulated results and significant findings below:
Posted by Eric Kraieski on Mon, Feb 15, 2010
Latest PUE Standards News from Green Grid
The Green Grid recently announced agreement by global world governments on using PUE as the primary measurement of data center efficiency:
"A group of global leaders met on February 2, 2010 to agree on data center energy efficiency measurements, metrics, and reporting conventions. Organizations represented were the U.S. Department of Energy's Save Now and Federal Energy Management Programs, U.S. Environmental Protection Agency's ENERGY STAR Program, European Commission JRC Code of Conduct, Japan's Ministry of Economy, Trade and Industry, Japan's Green IT Promotion Council, and The Green Grid."
"The collective groups are in agreement on the following guiding principles, as an interim step toward the desired outcomes (1. b.). It is recommended that data centers begin to measure PUE according to these principles... "
Here is the announcement from Green Grid:
Posted by Michael Bullock on Fri, Dec 11, 2009
Let's you and I do a little math about the National Security Agency's planned $1.5 billion data center.
Reading about the National Security Agency’s (NSA's) plan to build a $1.5 billion cyber security data center (Information Week) at the Camp Williams National Guard training base in Utah, I became curious about just how much one could glean about the project from public NSA budget information. Here's what I took away:
At first, this looks like a pretty good deal for Uncle Sam. We seem to be talking about a 1.5 million square foot data center that’s going to cost $1.5 billion, or $1,000 per square foot. That's well below what one would expect to pay for a Tier 3 class data center facility. Or is it?
For me, the math just doesn't add up. According to the budget document, the power density will be "appropriate for current state-of-the-art high-performance computing devices and associated hardware architecture." Yet if you calculate the watts per square foot by dividing the center’s total watts (65MW) by total square feet (1.5 million), you come up with a power density estimate of about 43 watts per square foot. No way that's "state of the art."
So let's say you triple the power density to a relatively modest 130 watts per square foot. That means you could support the center's full load of 65MW in about 500,000 square feet of space. At this power density, you'd probably require another 500,000 square feet for support space (generators, UPS, cooling, etc.). That's 1 million square feet. So what happens with the remaining half million square feet? According to the U.S. Army Corps of Engineers (as reported in the Information Week story) this data center "will eventually employ between 100 and 200 workers." That translates into a whopping 2,500 square feet of office space per employee. Roomy, wouldn't you say?
Now let’s suppose this facility really did support current state-of-the-art power density. That would mean about 400 watts per square foot. Given that density, the total space the data center actually would require would be just 160,000 square feet (i.e. 160,000 square feet x 400 watts per square foot = 64MW). Even assuming the same 500,000 square feet for support, that leaves us with a data center that's 660,000 square feet. Consequently, when you do the math, a 1.5 million square foot facility seems nutty and hard to reconcile with "state-of-the-art" anything. So what's going on?
My guess is that either the NSA has grossly miscalculated their space and power requirements or, perhaps, the true purpose and scale of the facility is a secret. The NSA keeping something secret? That wouldn’t be unheard of, would it?
Just for kicks, I wanted to see how this would compare to Uptime Institute estimates for this type of facility. I assumed 200 watts per square foot and used the Uptime Institute’s guidelines for a Tier 3 data center: $23,000 per KW of load and $300 per square foot. Using this approach, I came curiously close to NSA's $1.5 billion budget number ($1.495 billion, to be precise), and that made me wonder if the data center's budget is being built from the bottom up, or is simply a number tossed out to the public based on Uptime Institute estimates.
One last thought. Do you know how much money it will cost to operate a data center like the proposed Camp Williams facility? Based on a 65MW IT load, a PUE of 1.3 and Utah's cost of $.07 per kilowatt of electricity delivered (a nice rate while it lasts), it will cost $40 million per year simply to pay the electric bill.
But guess what? The reason the electricity is so cheap is because 98 percent of Utah’s electricity is produced by coal and natural gas. That's not very carbon friendly and with pending cap and trade legislation Utah's electricity costs will most definitely increase. How much? Who knows?
But whatever it is, the taxpayer (that's you) will be paying for it.
As always, I welcome feedback, questions and comments. You may reach me at cioblog@transitionaldata.com.
Posted by Michael Bullock on Mon, Jul 20, 2009
I’m going to stay away from the global warming and climate change debate; it looks (for now) like that train has left the station. Instead, I’d like to focus on the potential tax exposure created by "Cap and Trade” as it relates to the data center.
Last week I wrote about the Five Myths of Data Center Optimization, or Power Usage Effectiveness (PUE). However, a concept that is not a myth is that power distribution and cooling expenses act as a multiplier to the total cost to operate a data center. In other words, a data center operating at a 2.0 PUE level is generally wasting 1/3 more of the electricity it purchases to support IT operations than one operating at a more efficient PUE of 1.3. If you can buy power cheaply enough, it may seem like the penalty created by that 2.0 PUE rating is inconsequential. Trust me, it’s not.
If you look more closely, you’ll see that cheap electricity does not necessarily translate to a reduced carbon footprint. In fact, some electricity that’s cheap by today’s standards may increase substantially in total expense to you in the near future as Cap-and-Trade takes full effect. So, in the long run, making data center site selections based on today’s lowest cost of electricity may not be a wise strategy.
For example, the CO2 produced by nuclear and hydro-electric sources is minimal when compared to coal and petroleum-based plants, which are expected to produce 2.1 and 1.9 pounds of carbon dioxide gas respectively per kWh of electricity. Over time, as taxes work their way through the value chain and increase prices, you can expect the cost of electricity to reflect these differing methods of generation and their consequent emissions more accurately.
According to the EPA, the average emission rate of CO2 per kWh across all US regions is about 1.4 pounds. NY, CA, the Pacific Northwest and New England are all on the good side at about 0.9; the South averages around 1.5 and the Rockies come in at just over 2.0. And in the absence of some complicated, yet-to-be-defined measurement and auditing system, it seems logical that the federal government will use the EPA’s sub-region map for determining Cap and Trade allocations.
Here are two examples based on regional PUEs and local power grids:
• A data center in New England leveraging free cooling can achieve a sustainable PUE of about 1.3. At about .9 pounds of CO2, this facility generates about 1.17 pounds of CO2 per kWh.
• A data center in Texas with constant cooling demands will have a hefty PUE of about 2.0. Producing about 1.4 pounds of CO2 (per kWh), this facility will contribute 2.8 pounds of CO2 per kWh – for an increased carbon impact 140% above the New England example.
If you’re wondering just what impact this increased carbon footprint will have on global warming, join the club. There’s only one thing on which we can all probably agree – this increased carbon footprint will most assuredly be taxed. Bet on it. I don’t know about you, but avoiding an increased tax exposure of around 140% sounds like a pretty good idea to me.
By now you might be trying to figure out how cogeneration and alternative energy can help. Put down your pencil. Of the fossil fuel alternatives (eco-unfriendly, but reliable), natural gas is best at about 10 pounds per kWh (with diesel generators topping out over 22 pounds per kWh). Clearly, solar and wind power generation are carbon friendly, and may help your organization one of these days, but right now (and for the foreseeable future) they don’t scale to the magnitude and constant power demands required by corporate data centers.
I wish I could tell you how this will all play out in terms of specific tax impact, but your guess is as good as mine. In the meantime, the best advice is to focus on the stuff you can control. Optimize your PUE and don’t rush into a new facility without first taking a close look at your potential carbon exposure.
Listen carefully. Are those (carbon) footsteps you’re hearing? Or the tax man’s?
As always, I welcome your comments, tips, insights and topic suggestions. You can reach me at cioblog@transitionaldata.com.
Posted by Eric Kraieski on Wed, Jul 08, 2009
This is a "must read" for anyone who designs, builds, operates or maintains data centers.
The United States Environmental Protection Agency (EPA) developed this 2007 Report to Congress on Data Center Energy Efficiency in response to Public Law 109-431. This 130 page report assesses current trends in energy use and energy costs of data centers and servers in the U.S. and outlines existing and emerging opportunities for improved energy efficiency.
The 13 page Condensed EPA Report is also available.
Posted by Michael Bullock on Fri, Jun 26, 2009
Understanding and improving power usage effectiveness (PUE)
At a recent CIO / CFO conference, I saw why so many executives tune out when the topic turns to green IT and why they fail to make the "no-brainer" decisions that would immediately benefit their organizations. At the root of their apathy is the frequency with which they’re bombarded by green claims from vendors that often are misleading and sometimes absolutely meaningless.
As a reader commented on my recent blog Green IT Hype vs. Real Deal: "too many companies are running around hyping themselves as 'the green solution' while doing nothing more that re-advertising their same old products."
And this, my friends, is a big part of the problem. Despite valuable green innovations in the data center, the really good ideas are having trouble rising above the clatter of the vendors’ PR and sales machines.
Hopefully, data center executives can hear this: If you can improve your Power Usage Effectiveness (PUE) in ways that produce a fast payback, you will save money and reduce your carbon foot print. This is like hitting the Daily Double: You conserve natural resources while making money.
Think of it this way. What if you could instantly increase your car’s fuel efficiency by 35 percent via an upgrade that would pay for itself in the first year based on reduced fuel costs alone? You’d jump at the chance, right? So why is there so little interest in doing the same thing in data centers? Because it’s hard to tell which upgrades are real and which are only "the same old products" in new green wrapping.
I’m not going to dive into specific energy saving technologies; instead, I’m going to provide a framework that will allow you to get beyond the vendor hype and evaluate them on your own. (If you’re curious about what these technologies are, check out my recent CIO.com blogs on built green / built right and cool ways to save money).
A quick refresh: A data center’s PUE is the total load required to operate the IT systems plus support systems such as power distribution, humidification, and cooling. A PUE of 2.0 - which is typical of most data centers today - means that for every dollar spent on IT load you burn up another dollar on support. Some facilities have average annual PUEs over 3.0 (not good) and the best run data centers can sustain PUE levels around 1.3.
So here are five myths regarding PUE engineering that often get in the way of good data center optimization and design and prevent executives from making informed decisions for their data centers:
Myth 1 – PUE is a constant that applies 24/7/365
Not true. PUE can fluctuate by season, and even by time of day. For example, if the design leverages the free cooling of outside air, then that benefit will only be available when the outside temperature is optimal. That’s why it’s important to determine average annual PUE and not best case – which may only apply on cool days and nights. So when a vendor says that its technology will reduce your PUE to 1.3, take the time to find out if this claim is best case, steady state, or annual average.
Myth 2 - During data center design and construction, you should optimize PUE to full anticipated load.
Not only is this not true, it’s an approach that’s doomed from the start, almost guaranteed to leave you with an underutilized or oversubscribed data center for much of its life cycle. You’d be better served by thinking of your data center as a resource that will morph over time. This will guide you to make design decisions that emphasize flexibility.
For example, the power density in your data center will change over time and with it the demand for power and cooling. If you build for the target capacity, you’ll be spending money in the early years for more cooling capacity (and backup power) than you need. In other words, a half-full data center designed for an average annual PUE of 1.4 may run at 2.0 until it fills up, however long that may take.
Building a flexible environment means you might oversize pipes, raise your floor above three feet and use variable air handling options. In this way you can build a data center with a PUE that will provide more scalability (and efficiency) over an expected 10+-year life cycle.
Myth 3 – Alternative energy sources improve PUE
False. Supply side technologies such as cogeneration and renewable energy sources do not improve PUE, which is a measure of consumption efficiency. If anyone tells you that alternative energy sources will improve your PUE, keep asking them how they’ll improve power consumption efficiency. Eventually, they’ll shut up.
I’m not saying that renewable energy sources are bad; they simply don’t impact PUE. And sometimes they don’t save much money, either. Once maintenance costs are factored in, the power company will still need to provision power to you in case the system fails.
So you should evaluate these power sources on their own financial merit based on cost, reliability, and time-to-payback. Don’t jump in thinking these alternatives will earn you efficiency rebates or future carbon credits, especially if they’re driving inefficient power and cooling.
Myth 4 – You can’t improve PUE for existing facilities
Luckily, this is also false. If your data center is operating above its designed power and cooling capacity, there are things you can do to improve your PUE. The biggest gain can be achieved by removing heat-creating support systems from the data center floor. You can leverage outside air economizers; turn off the steam-based humidification systems, or replace CRACs with roof-top units. Anything that reduces heat generation or increases cooling can improve PUE, may have a fast payback, and can be subsidized through power company rebates. On the other hand, if you introduce a new technology that consumes electricity and adds heat to the system, you may be addressing one problem while creating another. For example, floor fan tiles are a sure giveaway that a facility is operating in a suboptimum range.
Myth 5 - New servers & virtualization improve PUE
This is partly true and partly false. By now, everyone should understand how server and storage virtualization can reduce the number of systems required to support your mission while lowering power consumption costs. Again, this alone does not improve PUE. If your data center is operating at the top 10 percent if capacity, simply removing load (i.e., heat generation) may reduce the duty cycle on the cooling system enough to improve PUE. But data centers operating below 80 percent of maximum capacity will not see PUE improvements through virtualization.
That said, these systems use less electricity and understanding how to leverage virtualization should be at the top of everyone’s green agenda.
Hopefully, dispelling these myths will allow you to ask the right questions of your architects, engineers, and vendors as you put together the data centers you’ll need to live with for 10 years or more. By evaluating PUE for yourself, you’ll be ready to save greenbacks and the environment.
Posted by Michael Bullock on Thu, Jun 04, 2009
I absolutely believe green is good. Unless it's being used in a misleading, manipulative or self-serving way.
I feel like the world is stuck in some kind of mash-up of the movies Wall Street and Groundhog Day. It’s an unsavory mix of Gordon Gekko (Michael Douglas’), “Greed is good!” mantra combined with weatherman Phil Connors’ (Bill Murray’s) loop of selfish, hedonistic habit.
In our real life version, we’re seeing entire industries cycling greed and going nowhere fast. Here is the pattern:
1. Find an angle to exploit.
2. Run it into the ground while ignoring the future.
3. At the end, when the check comes due, look for a handout.
Think about it. First we had the whole banking subprime mortgage fiasco. Once the government “modernized” the banking industry, practically everybody could borrow any amount of money they wanted based on the hype-fueled upward spiral of property values – which, of course, was a fantasy and not sustainable. And once the government stepped in to fix the problems it helped create, it became more difficult for qualified borrowers to get timely loans, making sure times stayed tough. This despite the fact that Bank of America and Citigroup received over $100 billion of taxpayer money combined (of the $200 billion banking total). (By the way, how much of that cash did you get? I’ve yet to receive my slice.)
And then we have GM which was loaned $15.4 billion by the US Treasury Department under the Troubled Assets Relief Program (TARP) in 2008 and then received another $30 billion this year. So where did those billions go? They circled down the toilet of debt service, payroll, severance, and other “black hole” expenses, and now they’re gone. Again, it’s a Groundhog Day loop with a centrifugal spiral of greed and habit.
I suppose I side with Texas Congressman Ron Paul who last November wrote, "In bailing out failing companies, they are confiscating money from productive members of the economy and giving it to failing ones. By sustaining companies with obsolete or unsustainable business models, the government prevents their resources from being liquidated and made available to other companies that can put them to better, more productive use.”
So where is the IT industry in this cycle? We’re definitely in the “find an angle to exploit” stage and that angle is “Green is good.” And you don’t need to buy into the whole global warming scenario to acknowledge that energy conservation and sustainability are worthy initiatives if for no other reason than that they conserve capital and preserve resources.
It certainly appears that everyone who sells goods, services and information to the IT sector has embraced green. Now don’t get me wrong. I absolutely believe green is good … unless it is being used in a misleading, manipulative or self-serving way.
For instance:
I recently attended the Network World, “IT Roadmap Conference and Expo” eager to hear what groundbreaking insights were going to be offered by Cisco and Verizon.
These companies sponsored a green IT seminar to discuss power management in the data center and office environments, utility computing, thin clients and virtualization as a green enabler. Unfortunately, what we got was an hour long infomercial on the merits of telecommuting and video conferencing. I felt like I had been suckered with a resort timeshare-type bait-and-switch.
There are, of course, good reasons to green IT. Virtualization clearly offers the promise of better resource utilization and lower operating expense. Energy-Star servers, outside air economizers and the like offer true green advantages. But, for example, are floor tiles with embedded fans green? No. In fact, they’re entirely unnecessary in a truly green facility. If you’re convinced you need them, I’m sure you’re wasting a lot more electricity than you realize. Is cogeneration green (i.e., producing your own primary electricity from fossil fuel-based generators)? While it may lower your electricity costs, it’s by no means carbon friendly and it’s certainly not green.
Here’s a question: When you think of Symantec, what comes to mind? Do you think “Green IT and data center optimization?” I don’t. Yet they’re on the green bandwagon: “Symantec helps IT decision makers to reduce energy consumption and increase space utilization by providing solutions that increase desktop, server, and storage efficiencies.” Ok, maybe this is a side benefit of their offering but it’s a bit of a stretch to call it green, no?
That said, Symantec has produced a decent white paper worthy of a peek: The Green Data Center – A Symantec Green IT Guide. Early in the report Symantec notes that half the power required for a data center is NOT used by the IT equipment; it’s used for power distribution and cooling. And then the remainder of the paper focuses on the IT side where the savings are more difficult to attain and quantify, and require a significant investment of time and money.
Why do so many companies, like Symantec, focus on the more difficult side of the equation? Because that’s the only side they can influence. It reminds me of the old adage: “To a hammer, everything looks like a nail.”
If you are running out of power, space, or cooling, you should be thinking about data center facility optimization because that’s where you’re likely to achieve substantial and immediate gains. When you hear the “green is good” drum beat, take the time to figure out what’s truly relevant. In many cases, you’ll find a lot of hype and very little substance.
So, are IT vendors are heading toward a Groundhog Day crash-and-bailout-and crash cycle? I don’t think so. This industry thrives on competition and innovation. With a relatively low level of unionized labor and a very small pension burden, the ground rules in tech are entirely different than they were for, say, GM. In the tech sector, boom and bust is as natural a cycle as cicadas and solar storms. We’ve all been there before. Companies offering innovation and substance will survive. Those that depend upon hype will perish. It has always been that way in tech and it always will be.
Posted by Michael Bullock on Mon, Mar 23, 2009
The role of PUE, DCiE and new performance metrics
According to the EPA’s 2007 report to Congress on server and data center energy efficiency,the two largest consumers of electricity in the data center are the support systems (50% of total) and general servers (34%). Last week I wrote about Energy Star Servers which over time will drive down the electrical demands of the servers. Today I’d like to focus on overall data center efficiency and productivity and the efforts of the Green Grid industry alliance.
For the most part, being green in the data center is about increasing power efficiency and thereby reducing greenhouse gas emissions. From a purely environmental and ecological perspective, higher levels of green could be achieved by examining the entire life cycle of technology systems— from production and supply chain all the way to the disposal of toxic components. But for now let’s stick with the power efficiency theme.
Here are some Green Grid developed metrics for evaluating data center facility efficiency:
- PUE—Power Usage Effectiveness represents how much total power you need to drive your IT systems when you take into account power distribution, cooling, humidification, lighting, etc. If you need 1MW to run your IT systems, a PUE of 1.8 means that you are consuming 1.8MW to get this usable level of power.
- DCiE—Data Center Infrastructure Efficiency represents the ratio of IT equipment power to total facility power. Using the same example, a facility with a PUE of 1.8 has a DCiE of about 55%, meaning a little more than half of the power used by the data center is making its way to the IT systems (no mystery here; DCiE is simply the inverse of PUE).
From a planning perspective, I’ve personally found PUE to be a more useful tool than DCiE because it gives you a simple multiplier you can use to estimate your increasing facility requirements as you add new servers, storage and the like. While PUEs generally fall into the 1.7 (more efficient) to 3.0 range (less efficient), I have seen PUEs approach 1.3 when a holistic, facility-wide approach is taken regarding power efficiency.
Keep in mind, the PUE and DCiE numbers tell you how efficiently your data center is operating from a power distribution and cooling perspective. They do not tell you how efficiently your IT group is delivering useful value per kW of total power consumed.
This is exactly what Green Grid is trying to do next: define ways to measure overall Data Center Productivity (DCP). That is, Green Grid is attempting to help you quantify how well the energy you’re using is being applied to useful work. Since this is very complicated and will vary by application and industry, the Grid is hoping to define useful “proxies” or indicators that can provide some normalized measurement and useful insight.
Here are some of the types of measurement proxies being considered by the Green Grid (read the Green Grid Proxy White Paper for more details):
- Server productivity.Because you need a greater number of older servers to get the job done, and because older servers consume a lot more electricity per instruction executed than newer ones, you could come up with a productivity measurement based on MCUPS (million compute units per second) per kW consumed by the data center (the more, the better). This, of course, ignores the value of stored data which may be considerable, or even required for regulatory compliance. However, it provides a measure worth consideration.
- Server utilization and virtualization. Running high-end servers at 20% capacity means they are being used inefficiently.The idea here is that running servers at higher utilization rates is good and a highly virtualized environment has the
- (server utilization continued) potential to produce more useful work per kWh consumed. Of course, this would not be good way to measure efficiency for applications that are seasonal or “bursty”— like those used by retailers (for, say, the Christmas rush), financial brokerages (with sudden floods of transactions) and even national security applications that are built for peak loads to guarantee performance in any eventuality.
- Bits out per kWh. The idea here is that it may be possible to measure useful work simply based on output. This may in fact be a great proxy for data intensive applications like websites, VoIP and video where the name of the game is moving bits efficiently. On the other hand, this would be a poor measurement for applications that aggregate and analyze data, then report out digestible, concise information, such as in business intelligence applications.
My purpose here is not to disparage the efforts of the Green Grid regarding DCP. I do believe, however, that the DCP models could use more input from the user community. That’s right: I mean you. Please post any comments here or, if you prefer, you can get involved with the Green Grid directly and provide feedback to the DCP survey.
At the same time, I encourage you to not lose sight that optimizing your PUE offers low hanging fruit in achieving improved efficiency, reducing carbon emissions and saving money. Modern facility technologies such as ultrasonic humidification, high efficiency harmonic mitigating transformers and variable frequency drives (VFDs) can be effective in dropping your PUE into the more efficient 1.3-1.5 range.
In fact, let’s compare IT server and storage upgrades with facility infrastructure improvements (PUE):
- For a baseline, let’s assume your IT equipment consumes 2MW with a PUE of 2.5. By investing in more efficient servers and storage, let’s say you are able to cut IT power demand by 30%. If the cooling and power systems scale down as well, you will cut your total power from 5MW to 3.5MW. Not bad— but also not really a quick fix as you migrate apps to your new servers and retire old systems.
- Now, using the same scenario, let’s say we improve facility efficiency instead (the support systems—cooling, humidification, distribution, etc.), driving PUE from 2.5 to 1.5 with changes that are invisible to the applications, servers and storage systems. In this scenario, you could reduce total electricity by 40% to 3MW and do it with a lot less migration expense and pain.
There’s no doubt that replacing older,inefficient computers and storage systems with more energy efficient models is a good idea. Clearly, ENERGY STAR servers and storage will help through natural technology refresh cycles. Virtualization will reduce the number of servers you need, allowing your systems to run at higher operational efficiency.
However, if you want to make an immediate impact that will continue to pay dividends as you refresh your servers, storage and network systems over time, improve your PUE first.
Posted by Michael Bullock on Fri, Mar 06, 2009
Data centers are widely recognized as energy hogs. By 2011, they are expected to account for 3 percent of all U.S. electricity consumption, according to the EPA. Not surprisingly, the government has identified the data center as a great place to start increasing American energy efficiency while reducing power consumption. Regardless of your political leanings, the writing is on the wall: Efficient IT will be rewarded and inefficient IT eventually will be penalized.
There are several reasons why the IT sector is being targeted by the government:
• It’s a quick fix. With technology refreshes occurring every three years on average, the IT industry is one of the few places where new technologies can be cut-in quickly. Contrast this to manufacturing, where machinery typically is expected to have at least a 10-year service life.
• It’s pervasive. All organizations except the very smallest use IT and have IT needs. IT is the one area that reaches across industries and government agencies of all sizes.
• It has the cash. The government believes the IT industry has deep pockets and the financial resources to invest. While this is classic big- government thinking, and it’s debatable as to whether having resources equates to an imperative to spend them, the ROI of energy efficiency is not at issue. After all, if you can buy a new server that can do the work of five older servers and pay for itself just in power savings alone, why not do it?
In any case, government agencies like EPA (Energy Star) and DOE (Data Center Energy Efficiency Program) are doing excellent work in promoting energy efficiency, as is the Green Grid consortium in the public sector.
Today I’d like to touch on the Energy Star server efficiency project. Since about half the electricity consumed in a data center is gobbled up by the IT equipment (the other half goes to cooling, humidification and other support services), it stands to reason that improving power efficiency at the server level would be a good thing.
While everyone knows about Energy Star personal computers, refrigerators, dishwashers, air conditions and the like, some are surprised to hear that there’s no equivalent standard for servers, storage and network gear. However, that’s about to change.
Two years in the making, Draft 4 of the Energy Star “Program Requirements for Computer Servers” was released February 20, 2009. The final draft is expected to arrive in early April and be approved around May 1. You can check out the details and progression (including vendor comments) by visiting the EPA Enterprise Server micro-site.
Here are some of the proposed criteria:
• Power supply efficiency. There will be minimum efficiency and power factor requirements at four load points: 10%, 20%, 50% and 100%.
• Idle power usage. To qualify for Energy Star, a computer’s idle power consumption must not exceed EPA’s defined maximum thresholds.
• Standardized information reporting. The manufacturer must provide access to standardized data sheets on their web site showing specific models and qualified Energy Star configurations.
• Power and temperature measurements. To qualify for the Energy Star label, a server must be able to provide real-time information on power consumption, inlet air temperature and utilization for all processor cores during normal operation.
As you can imagine, this can get pretty complicated and results in apples to oranges comparisons given the diversity of server types – from standalone 1U to high density rack servers. Consequently, the current Energy Star requirements have been reduced in scope to cover just 1-4 socket servers, which accounts for the majority of servers shipped on a unit basis. Rack servers will be covered in a later release of the specification.
So when can you expect to see the familiar Energy Star sticker begin popping up in your data center? Maybe never. Draft 4 eliminated the requirement that a product must carry the Energy Star label. There was some concern, it seems, that those decals might fall off and get sucked up into the fans.
So, yes, there are some kinks to be worked out, but over all this is good news for the industry. There’s money to be saved and efficiencies to be gained. In fact, next the EPA will be looking at the power efficiency of storage systems, with discussions beginning sometime this month or next.
Posted by Eric Kraieski on Mon, Mar 02, 2009
Read Green Data Center Article
View Green Data Center Slideshow
In February, Network World published a nice piece on the Internap data center in Sommerville, MA. Here are the first few paragraphs:
"Companies looking for a green data center model should take a look at the new facility Internap built in Somerville, Mass., just outside Boston, which is so environmentally efficient the local power company wrote it a rebate check for $453,000.
"A renovated warehouse building that most recently housed a 5,000-member church, the collocation facility is optimized to economically meet demands for cooling, humidity and power consumption that are common to all data centers.
"Internap expects that the data center will save another $400,000 every year by using less power than it would have had it not built to green specifications, says Mike Higgins, vice president and general manager of Internap's data center services. That is helpful to its bottom line as well as keeping down rates it charges customers, he says."