期刊文献+

基于Linux数据链路层MPI通信机制的设计与实现 被引量:1

A mechanism of MPI communication based on Data_Link_Layer
在线阅读 下载PDF
导出
摘要 针对MPI集群通信的特点,通过分析当前网络的通信结构和MPI的点到点通信模式,提出了一种基于数据链路层的集群通信机制,用以减少协议开销和内存拷贝次数,从而提高集群节点间的通信性能,并且通过实验验证了该机制的可行性。 In accordance with the feature of MPI cluster communication, a mechanism of MPI communicaton based on Data_Link_Layer is proposed in this paper in order to induce protocol cost and times of memory copy, by studying the communication structure in current network and MPI point-to-point communication models. This mechanism improves cluster inter-node communication efficiency. Finally, we demonstrate the effectiveness of this mechanism through experiments.
作者 王巍 李旺
出处 《电子技术应用》 北大核心 2012年第2期127-130,共4页 Application of Electronic Technique
关键词 内存映射 数据链路层 内存拷贝 memory map data link layer memory mopy
  • 相关文献

参考文献5

  • 1可向民,龚正虎,夏建东.零拷贝技术及其实现的研究[J].计算机工程与科学,2000,22(5):17-20. 被引量:17
  • 2刘路,谢旻,张磊,傅清朝.用户级通信中基于网络接口的虚实地址变换技术[J].计算机工程与科学,2008,30(9):154-157. 被引量:2
  • 3WILLIAM G, LUCK E, SKJELLUM A. Using MPI:portable parallel programming with the message passing interface[M]. 2nd Edition. Cambridge, MIT Press, 1999.
  • 4BUNTINAS D, GOGLIN B, GOODELL D,et al. Cache- efficient, intranode, large-message MPI communication with MPICH2-Nemesis[C].38th International Conference on Par- allel Processing(ICPP-2009). 2009:462-469.
  • 5毛德操 胡希明.Linux内核源代码情景分析[M].杭州:浙江大学出版社,2001..

二级参考文献12

  • 1[1] Bhoedjang R A F,et al.User-Level Network Interface Protocols[ J].IEEE Computer,1998,(11):53~60
  • 2[2] Walton S,Hutton A,Touch J.High-Speed Data Paths in Host-Based Ro uters[J].IEEE Computer,1998,(11):46~52
  • 3[3] Thadani M N,et al.A Efficient Zero-Copy I/O Framework for UNIX[ R].SML Technical Report,Sun Microsystems Inc,1995
  • 4[4] von Eicken T,Vogels W.Evolution of the Virtual Interface Architect ure[J].IEEE Computer,1998,(11):61~68
  • 5Steenkiste P. A Systematic Approach to Host Interface Design for High-Speed Networks[J]. IEEE Computers, 1994, 27(3) :47-57.
  • 6Schoinas I, Hill M D. Address Translation Mechanisms in Network Interfaces[C]//Proc of the 4th Int'l Symp on High Performance Computer Architecture, 1998 : 219-230.
  • 7Chen Y,Bilas A, Damianakis S N, et al. UTLB: A Mechanism for Address Translation on Network Interface[C] ff Proc of the 8th Int'l Conf on Architectural Support for Programming Languages and Operating Systems, 1998:193-203.
  • 8Lee Moon-Sang, Lee Sang-Kwon, Lee Joonwon,et al. Adopting System Call Based Address Translation into User-Level Communlcation[J]. IEEE Computer Architecture Letters, 2006,5(1) :7-7.
  • 9Li L, Vaisyanathan K, Trivedi K S. An Approach for Estimation of Software Aging in a Web Server[C]//Proe of the Int'l Syrup on Empirical Software Engineering, 2002:91- 100.
  • 10Tezuka H, O'Carroll F, Hori A,et al. Pin-Down Cache: A Virtual Memory Management Technique for Zero-Copy Communlcation[C]//Proc of 12th Int'l Parallel Processing Symp, 1998:308-314.

共引文献211

同被引文献14

  • 1JOHNSON A E, VOORHEES P W. A phase-field model for grain growth with trijunction drag [J]. Acta Materialia, 2014, 67:134-144.
  • 2BERG B A,WU H. Fortran code for SU (3) lattice gauge the- ory with and without MPI checkerboard parallelization [J]. Computer Physics Communications, 2012,183(10) : 2145-2157.
  • 3SUWA Y, SAITO Y, ONODERA H. Three-dimensional phase field simulation of the effect of anisotropy in grain-boundary mobility on growth kinetics and morphology of grain structure [J]. Computational Materials Science, 2007,40 (1) : 40-50.
  • 4SUWA Y,SAITO Y,ONODERA H. Phase field simulation of grain growth in three dimensional system containing finely dis- persed second-phase particles [J]. Seripta Materialia, 2006,55 (4) :407-410.
  • 5AMRITKAR A,TAFTI D, LIU R,et al. Open MP parallelism for fluid and fluid-particulate systems [J]. Parallel Computing, 2012,38(9) : 501-517.
  • 6DHOTE R,GOMEZ H,MELNIK R,etal. Isogeometric analy- sis of coupled thermo-meehanical phase-field models for shape memory alloys using distributed computing [J]. Procedia Com- puter Science, 2013,18:1068-1076.
  • 7WONG U H, AOKI T,WONG H C. Efficient magnetohydro- dynamic simulations on distributed multi-GPU systems using a novel GPU Direct-M_P/ hybrid approach [J]. Computer Physics Communications, 2014,185(7) : 1901-1913.
  • 8JIN H, JESPERSEN D, MEHROTRA P, et al. High perform- ance computing using MPI and OpenMP on multi-core paral- lel systems [J]. Parallel Computing, 2011,37(9) : 562-575.
  • 9朱昌盛,冯力,王智平,肖荣振.三维枝晶生长的相场法数值模拟研究[J].物理学报,2009,58(11):8055-8061. 被引量:6
  • 10卢照,张锦娟,师军,鱼佳欣.MPI动态负载平衡策略的研究与实现[J].计算机技术与发展,2010,20(5):132-135. 被引量:7

引证文献1

二级引证文献3

相关作者

内容加载中请稍等...

相关机构

内容加载中请稍等...

相关主题

内容加载中请稍等...

浏览历史

内容加载中请稍等...
;
使用帮助 返回顶部