Setting Standards: DC Measurement & Reporting

By agreeing on a common set of industry standards and guidelines, companies can more accurately monitor and measure the energy performance of their data center operations. Better measurement standards are the only way to create a baseline of consumption and effectively set targets for future use.


David Schirmacher is Vice President of 7x24 Exchange International and Chief Strategy Officer for FieldView Solutions and was one of the industry experts from across the industry who met on January 13, 2010 to agree on data center energy efficiency measurements, metrics, and reporting conventions.

The top issue that data center managers currently face is how to make their IT infrastructures more efficient and cost-effective. This is no easy task considering the volume of information the infrastructure is expected to handle.

For example, corporations, governments and other users created an estimated million gigabytes of information in 2009, according to the report “The Digital Universe Decade – Are You Ready?” by John Gantz and David Reinsel of IDC. The authors also report the digital data universe will explode to 35 zettabytes by 2020. Gartner Group agrees in its 2009 report indicating that enterprise data needs were expected to grow 650 percent through 2014.

As these infrastructures grow, so do associated energy costs. Gartner pegged energy-related costs as 12 percent of data center expenditures. This number is the fastest-growing cost associated with this infrastructure.

Energy Costs Continue to Climb

Controlling costs will be key for companies over the next decade. To effectively manage energy and power consumption, it’s necessary to have a clear understanding of energy metrics. For years, companies have tried to leverage disparate measurements to get an accurate picture of the energy efficiency of their data centers. However, these metrics are not always applied clearly and consistently. By agreeing on a common set of industry standards and guidelines, companies can more accurately monitor and measure the energy performance of their data center operations. Better measurement standards are the only way to create a baseline of consumption and effectively set targets for future use.

With this in mind, industry leaders (Data Center Metrics Task Force – comprised of top level executives of such leading organizations as: 7x24 Exchange, ASHRAE, The Green Grid, Silicon Valley Leadership Group, U.S. Department of Energy "Save Energy Now" Program, U.S. Environmental Protection Agency’s ENERGY STAR Program, United States Green Building Council, and the Uptime Institute) recently came to some agreement on data center energy efficiency measurements, metrics and reporting conventions.

The resulting recommendations are directly aimed at creating accurate and consistent measurement standards for reporting the overall energy efficiency of a data center. As reported in the group’s July 2010 whitepaper, “As business demands and energy costs for data centers rise, owners and operators have focused on the energy efficiency of the data center as a whole, frequently using energy efficiency metrics. However, the metrics are not always applied clearly and consistently.” (See PDF of Recommendations for Measuring and Reporting Overall Data Center Efficiency.)

Laying Out the Metrics

As a primary recommendation, all agreed that Power Usage Effectiveness (PUE) is the preferred metric for measuring the overall energy efficiency in a data center. PUE is a measurement of total data center energy consumption divided by the IT energy consumption. However, like most such metrics, PUE can produce misleading results – due to measurements taken at too short an interval, or at incorrect measurement locations.

At minimum, IT energy consumption should be measured at the output of the uninterruptable power supply (UPS). Additionally, whether dealing with a dedicated data center or one in a mixed-use building, the total energy consumption measurement must include all the energy required to operate the data center – including IT energy, cooling, lighting and support infrastructure.

Multiple Measurement Categories Provided

Taking this one step further, it’s important to establish a baseline for measuring energy consumption when calculating PUE. Ideally, to ensure all parties are operating at an accurate baseline, PUE should be measured over an aggregated 12-month period. Measurement over this longer time frame reduces the influence of temporary spikes or lulls in power consumption. Recognizing that not all data center operators have all of the necessary measurement equipment currently installed, the Task Force has provided several categories of measurement. These categories range from fairly simple periodic measurements that all operators should be able to implement immediately, to more sophisticated measurement strategies that provide greater accuracy.

As David Chernicoff in ZDnet wrote: “PUE is now broken out into four categories, 0 through 3, with 0 being a demand-based calculation that is most similar to the snapshot-like view of the original metric, while categories 1, 2, 3 are all consumption-based calculation based on actual usage measurements within the data center, with category 3 being the most detailed, consisting of 12-month total kWh measurements, taken at the point of connection between individual IT devices and the electrical system.”

So What Does This Mean In Terms Of Cost?

To be effective, companies must take the next step and translate energy usage into total costs to the business. To sell measurement as a viable cost and business issue, metrics must have a dollar value. This means that after the PUE benchmark is calculated, companies should determine the energy cost-savings by improving data center efficiency. Such metrics will translate raw numbers into business costs directly impacting the bottom line. This key next step is where many companies fail to execute.

As Michael Bullock states in his CIO blog: “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 footprint. This is like hitting the Daily Double. You conserve natural resources while making money.”

Challenges Continue

Data center and energy efficiency is no easy task. It is a challenging and complex process, especially for companies with no baseline metrics in place. The adoption of an industry-wide set of common metrics will help companies focus on measurement as an essential means to grasping an understanding of their power consumption while setting goals for future use. Once reliable industry-wide metrics are adopted and implemented, the industry can move forward with true energy efficiency. While it’s agreed that this is no simple feat, it would necessitate an ever-evolving model with room for continuous improvement.

Industry Perspectives is a content channel at Data Center Knowledge highlighting thought leadership in the data center arena. See our guidelines and submission process for information on participating. View previously published Industry Perspectives in our Knowledge Library.

Hide comments


  • Allowed HTML tags: <em> <strong> <blockquote> <br> <p>

Plain text

  • No HTML tags allowed.
  • Web page addresses and e-mail addresses turn into links automatically.
  • Lines and paragraphs break automatically.