Productive parallel programming with CHARM++.

HPC '15: Proceedings of the Symposium on High Performance Computing(2015)

引用 0|浏览14
暂无评分
摘要
CHARM++ is a general-purpose framework for developing high-performance parallel applications [1]. Applications written using Charm++ run at scales spanning mobile devices [2], multi-core processors, multi-processor NUMA woprkstations and servers, networked clusters, and the world's largest supercomputers. A selection of Charm++ applications in production scientific usage is shown in Table 1. These applications consume the second most execution cycles on NSF computing resources, after MPI.
更多
查看译文
关键词
productive parallel programming
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