Metrics and Measurements

White Paper

Return to Main

Comment Policy

The comment option is to provide feedback on this particular deliverable. Any advertisement of products will be removed.

WP#35-Water Usage Effectiveness (WUE™): A Green Grid Data Center Sustainability Metric

Posted In:

Metrics and Measurements



Michael Patterson, Intel

Dan Azevedo, Symantec
Christian Belady, Microsoft
Jack Pouchet, Emerson

TGG has developed a new metric to complement the series of metrics it has introduced in the past few years, which includes power usage effectiveness (PUE™), data center energy productivity (DCeP), energy reuse effectiveness (ERE), data center compute efficiency (DCcE), and others. The Green Grid proposes this new metric to address water usage in data centers, which is emerging as extremely important in the design, location, and operation of data centers in the future. The new water usage effectiveness (WUE™) metric—combined with the PUE and carbon usage effectiveness (CUE™) metrics—enables data center operators to quickly assess the water, energy, and carbon sustainability aspects of their data centers, compare the results, and determine if any energy efficiency and/or sustainability improvements need to be made. Since PUE has received broad adoption in the industry, the WUE metric is a natural extension of PUE and extends the family of xUE metrics.

an emerging metric that deserves careful attention

Thanks for developing this emerging efficiency metric for data centers. While the industry is moving to slash potable water consumption for cooling (e.g., using air economizer, recycled water, seawater, among others), very little attention has been paid by the research community to the enormity of water footprints of data centers (including both site water for cooling and source water for electricity generation)! I did some research on this emerging problem and proposed a new resource management solution to cutting data centers' water footprints. The basic idea is to "follow water-efficient data centers and time periods" by processing more workloads in more water-efficient data centers and/or time periods. While simple at a high level, there're many other technical challenges to address before my water-efficient resource management suite can be applied in real systems. However, my preliminary work has shown some promising results: by appropriately deciding "where" and "when" to process workloads, the data centers' water footprints can be effectively reduced by up to 50% whereas the energy consumption/electricity cost/carbon emissions remain almost unchanged. My solution adopts a software-based approach, which fundamentally differs from the current engineering-based methods (e.g., improving cooling systems).

Posted at 12:28 PM on October 27, 2013 by Shaolei Ren

Site issues come into play as well...

Justin: thanks for the comments, as you point out generally all of these things are a trade off. And energy over water may not always be the right choice... really depends on your locale as well... the choice may be water conservation in the southwest, but a wetter climate may choose energy. Each site needs to look at their balance of water, chemicals, carbon, energy, costs, and understand what is best for them.

Posted at 05:06 PM on February 21, 2012 by Michael K Patterson

Data Center Technician

The only problem is, all costs being equal, if it is a simple trade off between water and energy consumption, the question becomes, do I want to reduce water consumption and in turn increase my electricity use and therefore carbon emissions? The only answer, if I had to make it, at this point becomes a PR issue. which option makes me look better in the eyes of the public, and that, to me, would appear to be lower energy consumption regardless of the actual long term impacts of that decision.

Posted at 10:33 PM on February 10, 2012 by Justin Grau

Comment On This Document

Rate This


Thanks for visiting The Green Grid website! We are always working to improve our site and would appreciate your input.

Please click here to take a brief website survey.

No Thanks