Topology-aware GPU scheduling for learning workloads in cloud environments

作者：Marcelo Amaral, Jordà Polo, David Carrera, Seetharami Seelam, Małgorzata Steinder · 年份：2017 · DOI：10.1145/3126908.3126933 · 被引用次数：64 · 研究领域：Parallel Computing and Optimization Techniques、Distributed and Parallel Computing Systems、Interconnection Networks and Systems

Recent advances in hardware, such as systems with multiple GPUs and their availability in the cloud, are enabling deep learning in various domains including health care, autonomous vehicles, and Internet of Things. Multi-GPU systems exhibit complex connectivity among GPUs and between GPUs and CPUs. Workload schedulers must consider hardware topology and workload communication requirements in order to allocate CPU and GPU resources for optimal execution time and improved utilization in shared cloud environments.