期刊文献+

多计算机间的文件并行传输服务:MPI-TFTS

Files parallel transfer service among computers: MPI-TFTS
下载PDF
导出
摘要 在异构环境中,计算机间快速、安全、可靠的文件传输是非常必要,然而很多基于MPI的分布式并行开发环境没有提供计算机间的文件并行传输服务。提出了一种基于MPI的支持多种传输工具的文件并行传输服务:MPI-TFTS(message appli-cation interface tools file transfer service),MPI-TFTS采用三层体系结构,集成多种传输工具,使得其可以用MPI通信原语、现有的传输工具进行文件并行传输,在环境条件确定的情况下,用户可以选择更好的传输工具来减少文件传输时间,最后给出了两种文件传输服务的并行实施。 It is necessary for transferring files among computers quickly, securely, and reliably in the heterogeneous. However, many distributed and parallel environments based on MPI do not provide files parallel transferring services among multi-computers. MPI-TFTS (message application interface tools file transfer service) is proposed, which is based on MPI and can parallel transfer files among computers. MPI-TFTS integrates multi transfers tools using tree layers infrastructure method. So it can use MPI communicating principle or existent transfers tools to parallel transfer files. And users can select better transfers tools integrated in MPI-TFTS to decrease files transfers time if the transfer is not change. Finally, the parallel implement of two methods of file transfer service is given.
出处 《计算机工程与设计》 CSCD 北大核心 2006年第22期4207-4209,4270,共4页 Computer Engineering and Design
基金 国家自然科学基金项目(60072006)
关键词 体系结构 传输任务 文件传输服务 异构环境 并行传输 architecture transfer tasks file transfer service heterogeneous environment parallel transfer
  • 相关文献

参考文献8

  • 1Karonis N,Toonen B,Foster I.MPICH-G2:A grid-enabled implementation of the message passing interface[J].Journal of Parallel and Distributed Computing,2003,63(5):3-10.
  • 2Rashmi Bajaj,Dharma P Agrawal.Improving scheduling of tasks in a heterogeneous environment[J].IEEE Tran on Parallel and Distributed Systems,2004,15(2):107-108.
  • 3Cai M,Chervenak A,Frank M.A peer-to-peer replica location service based on a distributed Hash table[C].Washington,DC,USA:Proc of the ACM/IEEE SC 2004 Conf,IEEE Computer Society Press,2004:56-56.
  • 4William E Allcock,Ian Foster,Ravi Madduri.Reliable data transport:A critical service for the grid[J].Journal of Cluster Computing,2002,5(3):305-314.
  • 5Rajeev Thakur,William Gropp,Ewing Lusk.Data sieving and collective I/O in ROMIO[C].Washington,DC,USA:Appeared in Proc of the 7th Symposium on the Frontiers of Massively Parallel Computation,IEEE Computer Society Press,1999.182-189.
  • 6The LAM/MPI Team Open Systems Lab.LAM/MPI user's guide version 7.1.1[M].Trustees of Indiana University Press,2004.5-30.
  • 7Alain Roy,Ian Foster,William Gropp,et al.MPICH-GQ:Qualityof-service for message passing programs[C].Washington,DC,USA:Proceedings of the IEEE/ACM SC2000 Conference November,IEEE Computer Society Press,2000.1-11.
  • 8Matei Ripeanu,Ian Foster.A decentralized,adaptive,replica location mechanism[C].Edinburgh,Scotland:IEEE International Symposium on High Performance Distributed Computing,IEEE,2002.24-32.

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