中国科学院机构知识库网格
Chinese Academy of Sciences Institutional Repositories Grid
首页 机构 成果 学者
  
  1. CAS IR Grid
  2. 计算技术研究所
  3. 中国科学院计算技术研究所
  4. 中国科学院计算技术研究所期刊论文
  5. 英文
Cache-Oblivious MPI All-to-All Communications Based on Morton Order

文献类型：期刊论文

作者Li, Shigang2; Zhang, Yunquan2; Hoefler, Torsten1
刊名IEEE TRANSACTIONS ON PARALLEL AND DISTRIBUTED SYSTEMS
出版日期2018-03-01
卷号29期号:3页码:542-555
关键词cache-oblivious algorithms collective communication NUMA MPI_Alltoall MPI_Allgather neighborhood collectives
ISSN号1045-9219
DOI10.1109/TPDS.2017.2768413
英文摘要Many-core systems with a rapidly increasing number of cores pose a significant challenge to parallel applications to use their complex memory hierarchies efficiently. Many such applications rely on collective communications in performance-critical phases, which become a bottleneck if they are not optimized. We address this issue by proposing cache-oblivious algorithms for MPI_Alltoall, MPI_Allgather, and the MPI neighborhood collectives to exploit the data locality. To implement the cache-oblivious algorithms, we allocate the send and receive buffers on a shared heap and use Morton order to guide the memory copies. Our analysis shows that our algorithm for MPI_Alltoall is asymptotically optimal. We show an extension to our algorithms to minimize the communication distance on NUMA systems while maintaining optimality within each socket. We further demonstrate how the cache-oblivious algorithms can be applied to multi-node machines. Experiments are conducted on different many-core architectures. For MPI_Alltoall, our implementation achieves on average 1.40X speedup over the naive implementation based on shared heap for small and medium block sizes (less than 16 KB) on a Xeon Phi KNC, achieves on average 3.03X speedup over MVAPICH2 on a Xeon E7-8890, and achieves on average 2.23X speedup over MVAPICH2 on a 256-node Xeon E5-2680 cluster for block sizes less than 1 KB.
资助项目National Natural Science Foundation of China[61502450] ; National Natural Science Foundation of China[61432018] ; National Natural Science Foundation of China[61521092] ; National Key R&D Program of China[2016YFB0200800] ; National Key R&D Program of China[2017YFB0202302]
WOS研究方向Computer Science ; Engineering
语种英语
WOS记录号WOS:000425173200005
出版者IEEE COMPUTER SOC
源URL[http://119.78.100.204/handle/2XEOYT63/5655]  
专题中国科学院计算技术研究所期刊论文_英文
通讯作者Li, Shigang
作者单位1.ETH, Dept Comp Sci, CH-8092 Zurich, Switzerland
2.Chinese Acad Sci, Inst Comp Technol, State Key Lab Comp Architecture, Beijing 100864, Peoples R China
推荐引用方式
GB/T 7714		 Li, Shigang,Zhang, Yunquan,Hoefler, Torsten. Cache-Oblivious MPI All-to-All Communications Based on Morton Order[J]. IEEE TRANSACTIONS ON PARALLEL AND DISTRIBUTED SYSTEMS,2018,29(3):542-555.
APA Li, Shigang,Zhang, Yunquan,&Hoefler, Torsten.(2018).Cache-Oblivious MPI All-to-All Communications Based on Morton Order.IEEE TRANSACTIONS ON PARALLEL AND DISTRIBUTED SYSTEMS,29(3),542-555.
MLA Li, Shigang,et al."Cache-Oblivious MPI All-to-All Communications Based on Morton Order".IEEE TRANSACTIONS ON PARALLEL AND DISTRIBUTED SYSTEMS 29.3(2018):542-555.

入库方式： OAI收割

来源：计算技术研究所

浏览0
下载0
收藏0
其他版本

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。