arXiv:2410.11199cs.DCcs.AI2024-10被引 65

英国新建超算专攻AI训练,5448块GPU达21 ExaFLOP/s性能

Isambard-AI: a leadership class supercomputer optimised specifically for Artificial Intelligence

  • 采用HPE Cray EX4000架构,集成5448块NVIDIA Grace-Hopper GPU
  • 8-bit浮点算力超21 ExaFLOP/s,64-bit算力超250 PetaFLOP/s,功耗低于5MW
  • 支持云端式交互界面,便于研究人员快速上手开展AI实验

Isambard-AI是为支持人工智能研究而设计的新一代领导级超级计算机。基于HPE Cray EX4000系统,部署于英国布里斯托尔的新型节能模块化数据中心,配备5,448块NVIDIA Grace-Hopper GPU,可提供超过21 ExaFLOP/s的8位浮点算力用于大模型训练,以及超过250 PetaFLOP/s的64位浮点算力,整机功耗低于5MW。系统集成两个全闪存存储系统:20 PiByte的Cray ClusterStor和3.5 PiByte的VAST解决方案,支持训练、推理及安全数据共享。其软件栈面向使用云GPU的研究者,通过Jupyter笔记本、MLOps或网页交互界面访问,而非传统批处理作业方式。系统支持容器化,可快速迭代升级以适应快速演进的AI软件生态。一期将于2024年5月至6月上线,全年完成全部部署。

原文摘要 · Abstract (English)

Isambard-AI is a new, leadership-class supercomputer, designed to support AI-related research. Based on the HPE Cray EX4000 system, and housed in a new, energy efficient Modular Data Centre in Bristol, UK, Isambard-AI employs 5,448 NVIDIA Grace-Hopper GPUs to deliver over 21 ExaFLOP/s of 8-bit floating point performance for LLM training, and over 250 PetaFLOP/s of 64-bit performance, for under 5MW. Isambard-AI integrates two, all-flash storage systems: a 20 PiByte Cray ClusterStor and a 3.5 PiByte VAST solution. Combined these give Isambard-AI flexibility for training, inference and secure data accesses and sharing. But it is the software stack where Isambard-AI will be most different from traditional HPC systems. Isambard-AI is designed to support users who may have been using GPUs in the cloud, and so access will more typically be via Jupyter notebooks, MLOps, or other web-based, interactive interfaces, rather than the approach used on traditional supercomputers of sshing into a system before submitting jobs to a batch scheduler. Its stack is designed to be quickly and regularly upgraded to keep pace with the rapid evolution of AI software, with full support for containers. Phase 1 of Isambard-AI is due online in May/June 2024, with the full system expected in production by the end of the year.

超算AI训练GPU集群

Thank you to arXiv for use of its open access interoperability. PaperDance 不是 arXiv 官方产品;中文卡片由大模型生成,请以原文为准。