作者
Kiran Ranganath, Joshua D Suetterlein, Joseph B Manzano, Shuaiwen Leon Song, Daniel Wong
发表日期
2021/11/14
图书
Proceedings of the International Conference for High Performance Computing, Networking, Storage and Analysis
页码范围
1-14
简介
Multi-accelerator servers are increasingly being deployed in shared multi-tenant environments (such as in cloud data centers) in order to meet the demands of large-scale compute-intensive workloads. In addition, these accelerators are increasingly being inter-connected in complex topologies and workloads are exhibiting a wider variety of inter-accelerator communication patterns. However, existing allocation policies are ill-suited for these emerging use-cases. Specifically, this work identifies that multi-accelerator workloads are commonly fragmented leading to reduced bandwidth and increased latency for inter-accelerator communication.
We propose Multi-Accelerator Pattern Allocation (MAPA), a graph pattern mining approach towards providing generalized allocation support for allocating multi-accelerator workloads on multi-accelerator servers. We demonstrate that MAPA is able to improve the execution time …
学术搜索中的文章
K Ranganath, JD Suetterlein, JB Manzano, SL Song… - Proceedings of the International Conference for High …, 2021