期刊文献+

适用于工作站机群系统高性能计算的快速消息传递协议(英文) 被引量:2

Fast Message Passing for High Performance Computation in Workstation Clusters
下载PDF
导出
摘要 消息传递系统对于工作站机群系统的计算性能是至关重要的 .随着网络设备性能的飞速提高 ,传统的 TCP/IP协议中不必要的开销已经成为通信速度的瓶颈 .为了能够充分利用高速硬件的通信潜能 ,介绍了一种专门为Myrinet设计的新的通信协议—— FMP.该协议包含网络部分和本地部分 ,是一个精简的通信协议 .物理层假定是不会出错的 ,因此该协议十分简单而高效 .与 TCP/ IP相比 ,该协议的性能有了很大的提高 .介绍了该协议中的关键技术 ,并展示了部分结果 . The message passing system is crucial to the computational performance of workstation clusters. With the fast growth in performance of networking device, the unnecessary overhead in traditional TCP/IP protocol has become the bottleneck of communication speed. To fully exploit the communication potential of the fast hardware, this paper introduces FMP, a new communication protocol specifically designed for Myrinet. FMP, which consists of a network part, is a concise protocol. The physical layer is assumed to be error free and thus the protocol becomes extremely simple and highly efficient. Compared with TCP/IP, the performance of FMP is much better. In this article the essential techniques used in the protocol are discussed and some results are presented.
出处 《软件学报》 EI CSCD 北大核心 2001年第3期317-322,共6页 Journal of Software
基金 国家 8 6 3高科技发展计划基金&&
关键词 通信协议 消息传递协议 工作站机群系统 TCP/IP协议 FMP协议 Communication Network protocols Parallel processing systems
  • 相关文献

参考文献2

  • 1Dong C L,IEEE Computer Societyed Proceedings of the APDC'97,1997年,307页 被引量:1
  • 2Dubnicki C,Proc 1997 Int Parallel Processing Symposium,1997年,388页 被引量:1

同被引文献18

  • 1Ian Foster, Carl Kesselman. The Grid 2: B luep Rint for a New Computing Infrastructure[M]. San Francisco(CA, USA) : Morgan Kaufmann,2003. 被引量:1
  • 2Globus Alliance. GT4[J/OL]. E2005-06] http;//www.globustoolkit, org. 被引量:1
  • 3Ian Foster. Three Shades of Grid, Computerworld[M/CD]. [2005-04-25]. 被引量:1
  • 4杨中志.基于PVM的并行检查点系统的关键技术研究[D].武汉:华中科技大学,2001. 被引量:1
  • 5Steven Osman, Dinesh Subhraveti, Gong Su, et al. The Design and Implementation of Zap: A System for Migrating Computing Environments[R]. Proceedings of the 5th Symposium on Operating Systems Design and Implementation (OSDI 2002), Boston:[s. n.] ,2002:361-371. 被引量:1
  • 6Tannenbaum T, Litzkow M. The Condor Distributed Processing System[J]. Dr. Dobb's Journal, 1995,20 (2) :40-48. 被引量:1
  • 7CARL K, IAN F. The anatomy of the grid[EB/OL].http://www. globus. org/, 2002-01- 10. 被引量:1
  • 8FRANK S, VONWELCH. Grid security: the globus perspective [EB/OL]. http://www. globus. org/, 2005-01-16. 被引量:1
  • 9LIU Wenjie, GU Guochang. Security issues in grid environment[A]. Proceedings of IEEE International Conference on Services Computing[C]. Shanghai,China, 2004. 被引量:1
  • 10BIRMAN K, COOPER R, JOSEPH T A, et al. Isis-A distributed programming environment [M]. User ' s Guide and Reference Mannual, Cornell University,1996. 被引量:1

引证文献2

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