Energy Efficient Task Graph Execution Using Compute Unit Masking in GPUs

Marcus Chow,Kiran Ranganath, Robert Lerias, Mika Shanela Carodan,Daniel Wong

2021 IEEE/ACM Redefining Scalability for Diversely Heterogeneous Architectures Workshop (RSDHA)（2021）

引用 2|浏览14

暂无评分

摘要

The frontiers of Supercomputers are pushed by novel discrete accelerators. Accelerators such as GPUs are employed to enable faster execution of Machine Learning, Scientific and High-Performance Computing applications. However, it has been harder to gain increased parallelism in traditional workloads. This is why more focus has been into Task Graphs. AMD’s Directed Acyclic Graph Execution Engine (D...

查看译文

关键词

Runtime,Scalability,Software algorithms,Graphics processing units,Machine learning,Parallel processing,Supercomputers

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