期刊文献+

一种支持OpenMP线程绑定的分布式构件模型 被引量:4

A DISTRIBUTED COMPONENT MODEL WITH OPENMP THREAD AFFINITY SUPPORT
下载PDF
导出
摘要 针对分布式系统软件的计算性能和网络性能需求,分析适用于SMP结构的OpenMP并行编程方法,并提出一种支持并行计算的分布式构件模型。进一步针对多线程在操作系统调度下的无序迁移问题,分析OpenMP规范中线程绑定技术原理,设计计算构件中线程绑定接口的实现方案。采用计算与通信分离的思想,设计独立的计算与通信基本构件,并给出TCP连接子的具体设计方案。通过实验对比验证了线程绑定机制可以改善软件的并行性能及TCP连接子的可行性。 Aiming at the requirement of distributed system in computation and network performance,in the paper we analyse the OpenMP parallel programming method suitable for SMP structure and put forward a kind of distributed component model with parallel computing support.Moreover,aiming at the problem of unordered migration of multithreads in operating system scheduling,the rationale of thread affinity technique in OpenMP criterion is analysed,and the implementation scheme of its interface in computer component is designed as well.By separating the computation from communication,the independent computing component and communicating component are also designed,and the specific design scheme of the connector using TCP is represented.Through the experiments,we have compared and verified that the thread affinity mechanism is able to meliorate the parallel performance of software and the feasibility of TCP connector.
作者 王铮
机构地区 解放军
出处 《计算机应用与软件》 CSCD 北大核心 2013年第3期203-206,共4页 Computer Applications and Software
关键词 分布式系统 构件模型 SMP OPENMP 线程绑定 Distributed system Component model Symmetric multi-processor(SMP) OpenMP Thread affinity
  • 相关文献

参考文献7

二级参考文献20

  • 1Top 500 List[EB/OL]. [2008-11-01]. http://www.top500. org/lists/ 2008/11.
  • 2PGI User's Guide[EB/OL]. [2008-05-06]. http://www. nerse. gov/vendor_does/eray/pgi-ug71. pdf.
  • 3The GNU OpenMP Implementation[EB/OL]. [2009-05-07]. http://people. sc. fsu. edu/-burkardt/pdf/libgomp.pdf.
  • 4Thread Affinity Interface (Linux and Windows)[EB/OL]. [2009-04-05]. http://www. intel.com/software/products/ compilers/docs/fmac/doc files/source/extfile/optaps _ for/ common/optaps_openmp_thread_affinity.htm.
  • 5Whalen S. Optimizing the NPB CG Benchmark for Multi- Core AMD Opteron Microprocessors[M]. Cray Inc,2007.
  • 6Feng Huiyu, Rob F. Unstructured Adaptive(UA) NAS Parallel Benchmark, Version 1.0[R]. NASA Technical Report NAS-04-006,2004.
  • 7Lundberg L. Evaluating the Performance Implications of Binding Threads to Processors [C] // Proc of the 4th Int'l Conf on High-Performance Computing, 1997:393-400.
  • 8Chen Larry T,IEEE Communication Magazine,1997年,35卷,2期,62页
  • 9William D. Gropp. A user's view of OpenMP: The good, the bad, and the ugly. The Workshop on OpenMP Applications and Tools (WOMPAT 2000), San Diego, California, 2000.
  • 10G. Gazzaniga, P. Lanucara, et al. Rapid parallelization of the drift-diffusion model for semiconductor devices. The 4th Europcan Workshop on OpenMP (EWOMP 2002), Roma, Italy, 2002.

共引文献24

同被引文献40

  • 1卢世祥,韩松,王岩飞.合成孔径雷达实时成像转置存储器的两页式结构与实现[J].电子与信息学报,2005,27(8):1226-1228. 被引量:9
  • 2殷顺昌,赵克佳.一种基于POMP的OpenMP程序负载均衡分析方法[J].计算机工程与应用,2006,42(35):84-87. 被引量:3
  • 3Das S R, Fujimoto R, Panesar K S, et al. GTW: A time warp system for shared memory multiprocessors[C]//Proceedings of the Winter Simulation Conference. Lake Buena Vista, USA: IEEE Press, 1994: 1332-1339.
  • 4Chen J, Annavaram M, Du bois M. SlackSim: A platform for parallel simulations of CMPs on CMPs[J]. ACM SIGARCH Computer Architecture News, 2009, 37(2): 20-29.
  • 5Miller J E, Kasture H, Kurian G, et al. Graphite: A distributed parallel simulator for multieores[C]//Proceedings of the 16th IEEE International Symposium on High-Performance Computer Architecture. Bangalore, India: IEEE Press, 2010: 1-12.
  • 6DOU Yong, LEI Yuanwu, WU Guiming, et al. FPGA accelerating double quad-double high precision floating-point applications for exscale computing[C]//Proceedings of the 24th IEEE International Conference on Supercomputing. Tsukuba, Japan: IEEE Press, 2010: 325-336.
  • 7Greg S, Richard B, YANG Xiaoyun. Multicore image processing with OpenMP[J]. IEEE Signal Processing Magazine, 2010, 27(2) 134-138.
  • 8Bhattacharjee A, Contreras G, Martonosi M. Parallelization libraries: Characterizing and reducing overheads[J]. ACM Trans on Architecture and Code Optimization, 2011, 8(1): 5-29.
  • 9Schmidl D, Terbovan C, Mey D, et al. Binging Neste OpenMP programs on hierarchical memory architectures[R]. Tsukuba, Japan: Center for Computing and Communication of RWTH Aachen University, 2010: 5-8.
  • 10Akhter S,Roberts J.多核程序设计:通过软件多线程提升性能[M].李宝峰,富弘毅,李韬,译.北京:电子工业出版社,2007:167-170.

引证文献4

二级引证文献6

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