一句话：通过中压电源平面和优先级感知功耗封顶服务，实现数据中心25%以上功率超订并降低数亿美元成本。
随着云服务商持续扩展数据中心容量，功率超订成为降低 provisioning 成本的关键手段。本文基于 Google 规模部署经验，提出一种软硬件协同设计：一方面引入中压电源平面，使功率共享域可覆盖数十 MW 的设备；另一方面实现可扩展、快速且鲁棒的功耗封顶服务，在每个节点上协调多种优先级的工作负载。经过多年生产部署，该设计实现了 25% 或更高的功率超订，节省数亿美元数据中心成本，同时保持所有工作负载所需的可用性与性能。
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.
@inproceedings{paperbot5432,
title = {Data Center Power Oversubscription with a Medium Voltage Power Plane and Priority-Aware Capping},
author = {Varun Sakalkar and Vasileios Kontorinis and David Landhuis and Shaohong Li and Darren De Ronde and Thomas Blooming and Anand Ramesh and James Kennedy and Christopher Malone and Jimmy Clidaras and Parthasarathy Ranganathan},
booktitle = {ASPLOS},
year = {2020},
doi = {10.1145/3373376.3378533}
}