How To Benchmark Data Center Energy Costs
There are two ways to benchmark data center energy costs, writes Techmanic.com. The first, Corporate Average Datacenter Efficiency (CADE), was developed by the Uptime Institute and measures energy efficiency. The institute determines the facility efficiency by calculating the amount of space used and multiplying that by the energy efficiency of the building. The higher the CADE number, the more efficient the data center.
The other method is called Power Usage Effectiveness (PUE), developed by the Green Grid. The PUE is calculated by dividing the total utility load by the total IT equipment load. In this method, the lower the PUE the better.
Although both methods can be effective, the measurements needed to calculate the benchmarks can be difficult to obtain, warned Forrester analyst Doug Washburn. Larger businesses usually benefit from benchmarking because the bigger the data center the more energy is consumed and the more likely the company will have an incentive to reduce consumption, he said.
Christian Belady, principle potential and cooling architect for Microsoft Global Foundation Services
and Microsoft’s representative to the Green Grid, notes that there are limits to these metrics, especially for smaller businesses, but that even small organizations can benefit by benchmarking on a smaller scale.
Last year, Uptime Institute’s Executive Director Kenneth Brill wrote in Forbes that low Power Utilization Effectiveness numbers are “scientifically meaningless.”
Brill offers five tips on how to cut energy bills while increasing cooling reliability and stability.
Energy Manager News
- Making Solar Inverters Smarter
- Unlocking the Power of Building Data
- Lockheed Martin Installs the GridStar Storage System at Syracuse Facility
- Schneider Electric Unveils Continuous Efficiency
- Avista Lauds ‘Fair’ Settlement in Idaho Rate Case
- BGE’s SEED Program Offers Energy Discounts to 19 Commercial Customers
- Retailer Offers 100% Solar Plan in Texas
- Dissecting the Data Revolution