Uptime Institute Surveys Member Data Center Energy Practices
Over half of companies say that saving energy in in their datacenters is “very important” to them. Only four percent say it’s not important. But most of them don’t have any financial incentives to find the means to cut costs.
This is an issue that the Uptime Institute has long derided as a flawed approach that keeps datacenters from becoming as efficient as they should be. But in its 2012 Datacenter Industry Survey, the organization found in the Data Center Energy portion of the survey, that the problem persists. Just 20 percent of the 1,100 respondents to the survey say that the IT Department has responsibility for paying the energy bill for datacenters (71 percent said the bill went to the facilities/corporate real estate departments.) Uptime Institute says that the first wave of efforts to improve data center energy efficiency was led by facilities managers, not IT managers, for precisely this reason. (The survey covers other issues than data center energy, but being Green Computing Report, we’re just highlighting that part.)
In a bit of editorializing, the report points out that the reason IT managers should take charge is because they’re the ones who control the processes that need to change in order to improve energy efficiency. The most significant gains for the next wave of energy-efficiency innovation will be made at the application and data layers. That includes “consolidating applications and servers, de-duplicating data, removing comatose but power-draining servers, building redundancy into the applications and IT architecture rather than physical systems, and improving server utilization.”
Another finding is that different levels of commitment to making change seems to depend on where the companies are located. For example, 28 percent of the respondents in Europe say their IT departments pay the bills, and the report suggests that may be due to the fact that “Europe has considerable power costs and government regulation of environmental issues.” In Asia only 10 percent of the respondents’ IT departments paid the electricity bill.
When asked for the top two drivers for implementing energy efficiency, 82% of the respondents cited financial savings. The second biggest reason (47% of respondents) was recovering wasted capacity—which, as the Uptime Institute points out, can also save money.
Thirty-five percent said that their efforts were due to corporate social responsibility initiatives. Only 4% said it was due to pressure from customers.
And what about PUE? Since these numbers are self-reported, one has to always wonder how accurate they are. Twenty-nine percent say they don’t measure it at all. For the rest the measurement techniques used are as follows:
• PUE Category 0: IT load measured at UPS output(s). Total data center power measured at the utility meters. Peak utilization/demand in a single snapshot measurement: 15%
• PUE Category 1: IT load measured at UPS output(s). Total data center power measured at the utility meters. 12 Month cumulative readings: 20%
• PUE Category 2: IT load measured at PDUs supporting IT loads. Total data center power measured at the utility meters. Peak utilization/demand in a single snapshot measurement: 20%
• PUE Category 3: IT load measured at the point of connection of IT devices to electrical system. Total data center power measured at utility meters. 12 Month cumulative readings: 10%
• Use an alternative method for measuring PUE: 6%
The average reported PUE was in the 1.8 to 1.89 range.
There were differences between the way large and small organizations run their centers. The larger one tend to run warmer inlet air temperatures, use VFDs and use free cooling.
The most popular server supply air temperature range is 21 degrees C (48%) while cold- aisle containment is the most popular type of airflow containment used (44%).