Energy storage in data centers has mainly been used as devices to backupgenerators during power outages. Recently, there has been a growing interest inusing energy storage devices to actively shape power consumption in datacenters to reduce their skyrocketing electricity bills. In this paper, weconsider using energy storage in data centers for two applications in a jointfashion: reducing peak demand charges and enabling data centers to participatein regulation markets. We develop an optimization framework that captures thecost of electricity, degradation of energy storage devices, as well as thebenefit from regulation markets. Under this frame- work, using real dataMicrosoft data center traces and PJM regulation signals, we show theelectricity bill of a data center can be reduced by up to 20%. Furthermore, wedemonstrate that the saving from joint optimization can be even larger than thesum of individually optimizing each component. We quantify the particularaspects of data center load profiles that lead to this superlinear gain.Compared to prior works that consider using energy storage devices for eachsingle application alone, our results suggest that energy storage in datacenters can have much larger impacts than previously thought possible.
展开▼