Issue #1 | Summer 2022
Efficiency Metrics for Data Centers
Efficiency metrics have been developed to measure the efficiency of data center operations and to identify existing weaknesses in different areas. The respective metrics map different types of resource consumption. Power Usage Effectiveness (PUE) is one of the most widely used efficiency metrics. It measures the efficiency of energy use and the relationship between total energy consumption and that of IT infrastructure. With a PUE of 1, all the energy expended would flow into the infrastructure. A value of 2 would mean that cooling, lighting and the facility itself require just as much power as the IT infrastructure. Anything that doesn’t directly serve the operation of computing is considered non-infrastructure. In 2020, data centers in Germany had an average PUE of 1.63. The efficiency with which other resources are used can be measured in the same way. Water Usage Effectiveness (WUE) or Carbon Usage Effectiveness (CUE) record the amount of water consumed or carbon dioxide emitted for a fixed level of energy consumption. Many such efficiency metrics exist, also measuring things like the utilization of the technical infrastructure, for example, or the reuse of the waste heat generated.