Emerson Calls for Industrywide Metric for Measuring Data Center Efficiency

Emerson Network Power, a business of Emerson and the global leader in enabling Business-Critical ContinuityTM, thinks that must change and is today calling on the information technology industry to agree on a data center efficiency metric that plays the same role as the miles-per-gallon (MPG) metric for vehicle fuel efficiency. Emerson Network Power believes the industry metric should take into account not only energy used by data centers but also the amount of information processed by them.

“Our goal is to illustrate how a metric can uncover important insights about performance while helping data center managers prioritize efficiency efforts in ways they could not before,” said Jack Pouchet, director of energy initiatives for Emerson Network Power. “Ultimately, a universal metric will help data center professionals make significant improvements in energy efficiency while meeting growing performance demands.”

That’s important because U.S. Environmental Protection Agency (EPA) statistics show that data centers used 61 billion kilowatt hours of electricity in 2006. More urgently, the EPA estimates that data centers will consume 100 billion kilowatt hours of electricity in 2011.

Emerson Network Power offers these insights in a white paper it released today – Energy Logic: Calculating and Prioritizing Your Data Center IT Efficiency Actions. In the white paper, Emerson proposes an example metric that measures Compute Units Per Second (CUPS) Per Watt.

Using available data on servers and chips to measure data center energy usage and equipment performance, Emerson Network Power uncovered an important fact – while data center energy consumption is rising, the increases are overshadowed by dramatic gains in data output and efficiency. In other words, much more data is being processed now by a single Watt of electricity than in the past.

To promote industry discussion and debate toward developing an agreed-upon metric, Emerson Network Power offers three criteria an efficiency metric should meet:

* It should drive the right behavior by data center managers, enabling them to focus efforts on improving data output, energy efficiency, or both.
* Be available and published at the IT device level to help buyers make the right choices.
* Be scalable, from the IT device to the data center level.

Just as the MPG metric enables car buyers to decide which is the most fuel efficient model to buy, an industrywide, universally agreed upon metric will enable data center managers to compare the overall efficiency of data centers and focus on tactics that can improve under-performing centers. For example, a company with two data centers that have different results can identify why one center is operating less efficiently.

“We continue to emphasize that today’s increasingly efficient technologies, along with effective management strategies, can help drive data center efficiency,” said Michael Zatz, manager of the EPA’s ENERGY STAR Commercial Buildings Program. “Through continued discussion and information-sharing, the data center community can continue to deliver critical benefits to our economy and society while reducing energy usage.”

The new Emerson Network Power white paper advances a discussion started last year with its initial Energy Logic, a prioritized roadmap to optimizing data center energy use. The first Energy Logic analysis, while important to better understanding data center energy use without compromising availability or flexibility, did not address overall data center efficiency because, without a universally accepted metric for data center output, overall efficiency cannot be accurately quantified.

The white paper is available for download by visiting www.efficientdatacenters.com, a new Web site supported by Emerson Network Power to offer data center professionals information on the latest advances in energy efficiency. This new Web site features a simple calculator that data center professionals can use to calculate the efficiency of their own data centers.