Visit additional Tabor Communication Publications
April 04, 2008
The Green Grid is a relatively new organization focused on improving energy efficiency in datacenters and "business computing ecosystems." The potential relevance of this group's mission to HPC is obvious: machines at the high end of HPC installations are consuming megawatts of power to perform their calculations, and megawatts more to remove the heat they generate. On an energy per square foot basis, we have some of the most significant challenges in the business when it comes to provisioning IT gear.
As a community, however, we have only recently started to focus on our energy use. As fixed dollar procurements have allowed us to buy more and more FLOPS, and new machines are crammed into smaller and smaller packages, we've simply run out of capacity in existing power and cooling distribution systems. Even in facilities that have managed to keep pace with computing infrastructure demands, power bills that were once someone else's problem are showing up in our budgets with an alarming number of zeroes.
There are now pockets of groups in HPC -- especially in regions where power is priced at a premium or eco-awareness is at a fever pitch -- that are starting to take steps to reduce their energy footprint. They are trying to answer questions about the most efficient ways to deliver power (DC everywhere? 13kV to the compute floor?) and how to deliver just the right amount of cooling to precisely the right place. But these efforts are early, and a group like The Green Grid has promise as an entry point for HPC provisioners to get hooked in to the latest tips and techniques.
After nearly two years, The Green Grid now has over 150 members, including power companies, hardware vendors, and end user organizations. Their strategy is focused around three thrusts: data collection, data assessment, and technology proposals. These thrusts will help datacenters understand what data they should be collecting to get a handle on their energy performance (and how they should be collecting it), how to put that data in the context of best practices, and what technologies datacenter managers should be planning for as they look for ways to shrink their energy footprint.
The Green Grid got the ball rolling back in April of 2006 when the membership list consisted largely of AMD, Sun, HP and IBM, and membership was free (see, for example, the April 2006 article in ServerWatch). The group spent the next year drumming up members and creating an organization, and in February of 2007 announced that it had completed formation with the addition of many more companies, including key members Intel and Microsoft. The group also adopted a paid membership scheme: either $5,000 or $25,000 depending on whether you are committed to green IT, or just interested in it.
In August 2007 The Green Grid released its Technical Roadmap, and by October had released its second round of whitepapers: including one on proposed metrics for datacenter efficiency. Initially these weren't available to the general public, but the day following the release the group reconsidered the exclusivity of this important information and made the papers generally available (see reporting on the initial release at ZDNet, and follow-up reporting the second day following public release of the documents).
In February 2008 the group held its first technical forum in San Francisco. The event was held over two days, the first of which was open to the public and press. The second day was reserved for members to discuss work in progress, including a much-discussed peer review of the Lawrence Berkeley study on direct current in the datacenter. That study claimed that datacenters could save 10-20 percent on their energy costs by using DC power, but has faced some criticism about methodology. A peer review could do much to clear up those questions and provide clear direction for datacenters.
But to date this review, along with lots of other nuggets of what look to be useful information (like "Five Ways to Save Server Power" and "Seven Ways to Improve Cooling"), remains locked behind the membership wall. All of which has been the subject of strident criticism in the technology press. (See, for example, the TechTarget blog post by Matt Stansberry, and a recent article by Ashlee Vance at The Register.)
Despite the frequently negative reaction to the group's slower-than-desired progress, much of what you can find today at The Green Grid's Web site is genuinely useful. At the site you'll find white papers on proposed power efficiency metrics, studies of datacenter power distribution options, and guidelines for building energy efficient datacenters, all available to members and non-members alike.
Critics of the organization's lethargic formation and slow start have a minor point. There was a six month gap between the initial round of white papers and the Technical Roadmap release; the organization used this time to finalize the roadmap for public distribution. And it's also fair to say that they are going through growing pains, figuring out what should be released to the public and what needs to remain value-added for membership. But with no statutory authority, The Green Grid will only be relevant if it includes enough members to shape the direction of the industry, and for the time being it needs to take specific steps to grow that influence. And with some of IT's biggest companies sitting side by side, it is forgivable if the group experiences some friction as it goes through its developmental years.
As recently pointed out by Ken Oestreich of member company Cassatt Corp., there are other organizations already working the same issues: the Silicon Valley Leadership Group and its Energy Efficient Data Center Project, the US EPA's comprehensive study of Data Center Efficiency and its new work to create an EnergyStar rating for datacenters, and the datacenter assessment and rating tool being created by US Department of Energy's Save Energy Now program.
So, where is the real value to be found? Oestreich points out that The Green Grid is driven by its members, and those members right now are heavily weighted toward technology companies. It follows that they have a real opportunity to work with regulatory groups like the EPA and DOE to catalyze efforts in those communities around standards and best practices, and to provide a ready source of focused feedback on new proposals. The Green Grid also has a real opportunity to take a lead in educating the end user community about the range of energy-saving technologies and practices developed across the industry, and provide objective assessments to help users in taking meaningful steps to improve the effectiveness of their datacenters. Although given their vendor-weighted membership, they'll have to be careful to remain, and be seen to be, objective.
Oestreich also suggests that The Green Grid can make a difference in proposing new, non-incremental energy-friendly technologies and getting those adopted in the broader community. I disagree with this point on the basis that The Green Grid is dominated by technology companies whose bottom lines are driven by the uniqueness of their products. Who would voluntarily give up a market lead in a new technology to bolster the collective relevance of a trade organization while, at the same time, giving that technology to their competitors?
The Green Grid is experiencing tremendous growth and is only just reaching two years of active operations. It takes time to build broad-based industry coalitions, just as it will take time to develop the body of work necessary to answer many of the fundamental questions IT provisioners are facing right now. Some of the criticisms aimed at the organization to date have merit, but none are fatal to their mission.
The influence on computational infrastructure on operational and even acquisition decisions in HPC will no doubt continue to grow in the very near future, and The Green Grid represents a key connection point for our community to engage with the businesses and fellow travelers in IT, who, despite their differences in mission space, ultimately share many of the same infrastructure concerns we do.
May 16, 2013 |
When it comes to cloud, long distances mean unacceptably high latencies. Researchers from the University of Bonn in Germany examined those latency issues of doing CFD modeling in the cloud by utilizing a common CFD and its utilization in HPC instance types including both CPU and GPU cores of Amazon EC2.
May 15, 2013 |
Supercomputers at the Department of Energy’s National Energy Research Scientific Computing Center (NERSC) have worked on important computational problems such as collapse of the atomic state, the optimization of chemical catalysts, and now modeling popping bubbles.
May 10, 2013 |
Program provides cash awards up to $10,000 for the best open-source end-user applications deployed on 100G network.
May 09, 2013 |
The Japanese government has revealed its plans to best its previous K Computer efforts with what they hope will be the first exascale system...
May 08, 2013 |
For engineers looking to leverage high-performance computing, the accessibility of a cloud-based approach is a powerful draw, but there are costs that may not be readily apparent.
05/10/2013 | Cleversafe, Cray, DDN, NetApp, & Panasas | From Wall Street to Hollywood, drug discovery to homeland security, companies and organizations of all sizes and stripes are coming face to face with the challenges – and opportunities – afforded by Big Data. Before anyone can utilize these extraordinary data repositories, however, they must first harness and manage their data stores, and do so utilizing technologies that underscore affordability, security, and scalability.
04/15/2013 | Bull | “50% of HPC users say their largest jobs scale to 120 cores or less.” How about yours? Are your codes ready to take advantage of today’s and tomorrow’s ultra-parallel HPC systems? Download this White Paper by Analysts Intersect360 Research to see what Bull and Intel’s Center for Excellence in Parallel Programming can do for your codes.
In this demonstration of SGI DMF ZeroWatt disk solution, Dr. Eng Lim Goh, SGI CTO, discusses a function of SGI DMF software to reduce costs and power consumption in an exascale (Big Data) storage datacenter.
The Cray CS300-AC cluster supercomputer offers energy efficient, air-cooled design based on modular, industry-standard platforms featuring the latest processor and network technologies and a wide range of datacenter cooling requirements.