Accelerating Distributed Machine Learning in Disaggregated Architectures with Flexible Optically Interconnected Computing Resources
2022 OPTICAL FIBER COMMUNICATIONS CONFERENCE AND EXHIBITION (OFC)(2022)
摘要
We introduce an optically interconnected disaggregated architecture for GPU resources and demonstrate a 3 x increase in GPU utilization and up to 73.2% acceleration of application runtime for distributed machine learning workloads. (C) 2022 The Author(s)
更多查看译文
关键词
GPU resources,GPU utilization,distributed machine learning workloads,optically interconnected disaggregated architecture,optically interconnected computing resources
AI 理解论文
溯源树
样例
生成溯源树,研究论文发展脉络
Chat Paper
正在生成论文摘要