Data Center Power Oversubscription with a Medium Voltage Power Plane and Priority-Aware Capping

Session: Datacenter/cloud power/performance--Managing the beast.

Authors: Varun Sakalkar (Google LLC); Vasileios Kontorinis (Google LLC); David Landhuis (Google LLC); Shaohong Li (Google LLC); Darren De Ronde (Google LLC); Thomas Blooming (Google LLC); James Kennedy (Google LLC); Christopher Malone (Google LLC); Parthasarathy Ranganathan (Google LLC)

As major web and cloud service providers continue to accelerate the demand for new data center capacity worldwide, the importance of power oversubscription as a lever to reduce provisioning costs has never been greater. Building on insights from Google-scale deployments, we design and deploy a new architecture across hardware and software to improve power oversubscription significantly. Our design includes (1) a new {\em medium voltage power plane} to enable larger power sharing domains (across tens of MW of equipment) and (2) a {\em scalable, fast, and robust power capping service} coordinating multiple priorities of workload on every node. Over several years of production deployment, our co-design has enabled {\em power oversubscription of 25\% or higher}, saving hundreds of millions of dollars of data center costs, while preserving the desired availability and performance of all workloads.