A dashboard screen shot from Power Assure’s EM/4 platform shows both facility energy efficiency and IT energy efficiency
(WEB HOST INDUSTRY REVIEW) — Data center optimization software developer Power Assure (www.powerassure.com) published a collection of insights on data center energy efficiency rating systems on Wednesday.
Clemes Pfeiffer, Power Assure’s chief technology officer, says the popular Power Usage Effectiveness rating system and the Data Center Infrastructure Efficiency rating have brought awareness to the importance of measuring data center efficiency, however he thinks they do not provide any criteria to show incremental improvements in efficiency over time.
In addition, according to Pfeiffer, neither PUE or DCiE monitor effective use of power supplied.
Pfeiffer says broader rating systems created by McKinsey and Garnet have done a better job of accounting for IT energy efficiency, including the Corporate Average Data Center Efficiency and Power to Performance Effectiveness systems.
Developed by Gartner, PPE identifies at the device-level where efficiencies can be gained. It does this by allowing users to define their own optimal maximum performance levels, and compare average performance against the optimum.
McKinsey’s CADE allows organizations to visualize and manage the efficiency, capacity, and performance of data centers by addressing poor server utilization.
“Power consumption in the data center is on the rise as demand for Internet services like music and video downloads increase and the number of servers needed to deliver this data multiply and age,” Pfeiffer said in a statement. “Visibility needed to manage operational and design improvements in the data center are spurring the adoption of energy metrics along with data center infrastructure management software that can establish the power baseline, and can then track improvements using multiple rating systems.”
EnergyStar is a rating system that helps IT managers more accurately determine when to refresh servers. Pfeiffer sees what he calls a “fundamental flaw” similar to the one found in PUE and DCiE systems that it ignores transactional efficiency of servers.
The final rating system was created by Power Assure and is a method for determining absolute and normalized energy efficiency ratings for equipment on a transaction-per-second per-watt basis. The system measures power-on spike, wave form, boot cycle and 100 percent load.
Power Assure also published a whitepaper called “Combining PUE with other energy efficiency metrics” which provides more detailed statistics of energy efficiency in data centers, as well as the specific equations used to determine efficiency.
No related posts.











