期刊文献+
任意字段
题名或关键词
题名
关键词
文摘
作者
第一作者
机构
刊名
分类号
参考文献
作者简介
基金资助
栏目信息

多核机群下基于神经网络的MPI运行时参数优化 被引量：3

MPI Runtime Parameters Tuning Based on Neural Network on Multi-core Clusters
在线阅读 下载PDF
导出
摘要 多核处理器的新特性给MPI应用带来了新的优化空间,其中调优MPI运行时参数被证明是优化MPI应用的有效方法。然而最优的运行时参数不仅与多核机群的体系结构有关,也决定于MPI应用的程序特征。提出并分析了一种在给定多核机群下基于人工神经网络的优化模型,用于自动为未知的MPI程序预测接近最优的运行时参数。两个不同基准的实验证明了本方法的有效性。实验证明,基于本方法得到的运行时参数所产生的加速比平均达到了实际最大加速比的95%以上。 The new features of multi-core add the optimization space for MPI applications, and besides tuning MPI runtime parameters is a common practice perceived to optimize the MPI application performance. However, the best configuration of the runtime parameters not only depends on the underlying architecture of a specific multi-core cluster but also on the features of MPI application. We constructed and analyzed an effective tuning model bases on artificial neural network to automatically predict the near-optimal configuration of runtime parameters for any unseen input programs under the current multi-core cluster. Experimental results from two different benchmarks were presented to show effectiveness of our approach. We observed that the speedup gained by the predicted runtime parameters can averagely achieve 95% of the speedup gained by the best parameters configuration.
作者 王洁 曾宇 张建林
机构地区 首都师范大学信息工程学院 中国科学院计算技术研究所 奥地利因斯布鲁克大学分布式与并行计算研究实验室 北京计算中心
出处 《计算机科学》 CSCD 北大核心 2010年第6期229-232,共4页 Computer Science
基金 奥地利蒂罗尔州未来基金会基金(P7030-015-024)资助
关键词 多核机群 MPI 运行时参数优化 神经网络 Multi-core clusters,MPI,Runtime parameters tuning, Neural network
分类号 TP393.09 [自动化与计算机技术—计算机应用技术]
  • 相关文献

参考文献10

  • 1Chai L, Lai P,Jin H W, et al. Designing an efficient kernel-level and user-level hybrid approach for MPI intra-node communication on multi-core systems[C]//ICPP ' 08:Proceedings of the 2008 37th International Conference on Parallel Processing. Washington, DC, USA, IEEE Computer Society, 2008.
  • 2TOP 500Team. TOP500ReportforJune2009 [EB/OL]. http:// www. top500, org.
  • 3Chaarawi M, Squyres J M, Gabriel E, et al. A tool for optimizing runtime parameters of Open MPI[C]//Proceedings of the 15th European PVM/MPI Users" Group Meeting on Recent Advances in Parallel Virtual Machine and Message Passing Interface. Berlin, Heidelberg, Springer-Verlag, 2008 :210-217.
  • 4Pellegrini S, WangJie, Fahringer T, et al. Optimizing MPI Runtime Parameter Settings by Using Machine Leaming[C] //Proceedings of the 16th Euro PVM/MPI Users' Group Meeting on Recent Advances in Parallel Virtual Machine and Message Passing Interface. Espoo, Finland, 2009.
  • 5Ipek E, Supinski B R, Schulz M, et al. An Approach to Performance Prediction for Parallel Applications[C]//Euro-Par Parallel Processing. Monte de Caparica, Portugal, August 2005.
  • 6Lee B C,Brooks D M,de Supinski B,et al. Methods of inference and learning for performance modeling of parallel applications [C] // PPoPP: 12th Symposium on Principles and Practice of Parallel Programming. San Jose,CA,March 2007.
  • 7Pjesivac-Grbovic J, Fagg G E, Bosilca G, et al. Decision Trees and MPI Collective Algorithm Selection Problem [C] // Euro-PAR 2007. LNCS 41541,2007.
  • 8Duan Rubing, Nadeem F, Wang Jie, et al. A hybrid intelligent approach for performance modeling and prediction of workflow activities in Grids[C]//9th International Symposium on Cluster Computing and the Grid. IEEE Computer Society, Shanghai, China, 2009.
  • 9Battiti R First-and second-order methods for learning between steepest descent and Newton's method[J]. Neural Computation, 1992,4(2) : 141-166.
  • 10Matlab: Neural Network Toolbox[ EB/OL]. http://www, mathworks,com/produets/neuralnet/.

同被引文献22

  • 1朱辉杰.多核硬件技术与通信集合之间的关系与应用.制造技术,2009,(11).
  • 2陈晓鹏.高性能计算与冲集合通信相结合.中国计算机用户,2009,(29).
  • 3王忠汇.多核时代的高性能数据处理与集合通信技术.新技术新工艺,2007,(4).
  • 4Allen J R,Kennedy K.现代体系结构的优化编译器[M].张兆庆,乔如良,冯小兵,等,译.北京:机械工业出版社,2004:23—90.
  • 5Bulic P,Gustin V. On Dependence Analysis for SIMD Enhanced Processors[J]. Lecture Notes in Computer Science, 2005,3402 (1) :527-540.
  • 6Bulic P,Gustin V. D-Test: An Extension to Banerjee Test for a Fast Dependence Analysis in a Multimedia Vectorizing Compiler [-C]/,/Proceedings of IPDPS, 2004. Washington 1312: IEEE Com- puter Society 2004.
  • 7Venkatasubramanyam R D. Array Access Analysis in Open64 I-D]. Houston: University of Houston, 2004.
  • 8Pouzols F M, Lendasse A, Barros A B. Autoregressive time se- ries prediction by means of fuzzy inference systems using non- parametric residual variance estimationl-J-]. Fuzzy Sets and Sys- tems, 2010,161(4) : 471-497.
  • 9Zhou Jing, Zeng Guo-sun. A general data dependence analysis for parallelizing compilers[J]. The Journal of Supercomputing, 2008,45 (2) : 236-252.
  • 10Drakakis K,Gow R, Rickard S. Distance Vectors in Costas Ar- raysrC]//Proceedings of the 42nd Annual Conference on Infor- mation Sciences and Systems. Washington I)C: IEEE Computer Society Press, 2009.

引证文献3

计算机科学
2010年 第6期

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