The electric power supply for AI data centers is now the most significant bottleneck in the race toward Artificial General Intelligence, surpassing even the constraint of AI accelerator availability. To our knowledge, this paper is the first to describe the end-to-end power management process for a hyper-scale AI datacenter; from early power planning to accommodate next-generation accelerators 6--12 months before their general availability, to tuning power settings after large scale deployment, and finally to dynamic, runtime power management for evolving workloads. We present detailed power measurements for a 150 MW datacenter hosting a cluster of 83K GB200 GPUs. We share insights from building this state-of-the-art AI cluster. We hope this work encourages practitioners across the industry to share their own experiences as well.

翻译：人工智能数据中心的电力供应已成为实现通用人工智能（AGI）的最大瓶颈，甚至超过AI加速器可用性的约束。据我们所知，本文首次描述了超大规模AI数据中心的端到端供电管理流程：从为下一代加速器实现提前6-12个月的早期电力规划，到大规模部署后的功耗设置调优，直至针对动态演变的运行负载进行运行时电力管理。我们展示了容纳83K GB200 GPU集群的150 MW数据中心的详细功耗测量数据，并分享了构建该先进AI集群的实践经验。希望这项研究能激励业界从业者共同分享自己的实践经验。