Okay, so around our home, especially when my daughter’s home from TCU, you are more likely to hear –“Hey, where did all the HOT water go?” But when the water bill comes in and after my breathing returns to normal we have a serious family discussion about conserving precious resources.
And it’s time we do the same with our data centers, which use a lot of water for cooling. Not as much water as my daughter, but a lot. That’s why today, I’d like you to start thinking about your data center’s WUE. Our WOO? No, that is Water Usage Effectiveness. Think of it as your water productivity metric. I haven’t single-handily invented a new data center metric but do suggest this is a topic whose time has come.
Why water? When more than 1 billion people lack access to safe drinking water and fully one-third of the world lives in “water stressed” countries, it is time for us to seriously rethink our use of water within our data centers. Water is a life necessity that we must carefully guard and protect in order to ensure our way of life.
So consider for a moment the ideas of Power Usage Effectiveness, aka PUE, as defined by The Green Grid and the general concept of Productivity, especially as they relate to business – where tangible outputs are related to a unit of input. I often think of earnings per share (EPS), sales per employee, sales per fixed assets, along with a host of other financial productivity metrics used to gauge the value of any for-profit business.
Now let’s take that idea and tie it to water. Why? Well, many people feel that water is the next oil and though that analogy is probably not the best because of the life-giving value of water, it certainly helps us understand the potential relevance as a unit of measure in evaluating productivity.
Therefore I strongly encourage the data center industry to formulate a water usage effectiveness WUE, or better yet a Water Systems Productivity (WSP) metric. Simply put, for a data center, take useful work or even a proxy for useful work such as Emerson’s proposed Compute Units Per Second and divide that by the amount of water used during the observation period. Water will most be likely measured in units, with 1 unit equal to an acre/foot. It’s a measure often used in the residential/commercial real estate sector --think of it as the amount of water required to flood one acre to the depth of one foot. Of course, gallons/liters work just fine if you choose.
This Water Systems Productivity metric would ideally be reported monthly along with all your other productivity and efficiency metrics. Once we start to measure and report water utilization, we will quickly realize that simply flowing more cooling water in order to “economize” may not always be the best answer. Now we will be able to have a meaningful tool to determine the ideal mix between dry-coolers, CW plants, and free-flow water-cooled systems, etc. versus the increased energy used with alternative solutions.
Living in Southern California, I now have a keen understanding of the importance of protecting our very limited and often questionable supply of precious drinking water. But even if you live in my former Olympic Peninsula home town of Forks, WA (note: I lived there before the vampires moved in) where we measured rainfall in FEET not inches (Yes, that was 12 feet in a year),we all must take our business use of water more seriously. Because the last time I looked they weren’t making any more.
Let’s help the EPA, EU, Green Grid, and others drive a meaningful water productivity metric.
Tags: pue,
power usage effectiveness,
the green grid,
compute units per second,
cups,
data centers,
water cooling,
free-flow water cooled systems,
epa,
eu
Categories: