MPI+Threads: Runtime Contention and Remedies | |
Abdelhalim Amer; Huiwei Lu; Yanjie Wei; Pavan Balaji; Satoshi Matsuoka. | |
2015 | |
会议名称 | 20th ACM SIGPLAN Symposium on Principles and Practice of Parallel Programming, PPoPP 2015 |
会议地点 | 三藩市 |
英文摘要 | Hybrid MPI+Threads programming has emerged as an alternative model to the “MPI everywhere” model to better handle the increasing core density in cluster nodes. While the MPI standard allows multithreaded concurrent communication, such flexibility comes with the cost of maintaining thread safety within the MPI implementation, typically implemented using critical sections. In contrast to previous works that studied the importance of critical-section granularity in MPI implementations, in this paper we investigate the implication of critical-section arbitration on communication performance. We first analyze the MPI runtime when multithreaded concurrent communication takes place on hierarchical memory systems. Our results indicate that the mutex-based approach that most MPI implementations use today can incur performance penalties due to unfair arbitration. We then present methods to mitigate these penalties with a first-come, first-served arbitration and a priority locking scheme that favors threads doing useful work. Through evaluations using several benchmarks and applications, we demonstrate up to 5-fold improvement in performance. |
收录类别 | EI |
语种 | 英语 |
内容类型 | 会议论文 |
源URL | [http://ir.siat.ac.cn:8080/handle/172644/6969] |
专题 | 深圳先进技术研究院_数字所 |
作者单位 | 2015 |
推荐引用方式 GB/T 7714 | Abdelhalim Amer,Huiwei Lu,Yanjie Wei,et al. MPI+Threads: Runtime Contention and Remedies[C]. 见:20th ACM SIGPLAN Symposium on Principles and Practice of Parallel Programming, PPoPP 2015. 三藩市. |
个性服务 |
查看访问统计 |
相关权益政策 |
暂无数据 |
收藏/分享 |
除非特别说明,本系统中所有内容都受版权保护,并保留所有权利。
修改评论