计算机科学 ›› 2014, Vol. 41 ›› Issue (6): 22-26.doi: 10.11896/j.issn.1002-137X.2014.06.005
张拥军,林宇斐
ZHANG Yong-jun and LIN Yu-fei
摘要: 随着并行计算系统规模的增大和复杂度的提高,已有的多作业分配方式可能导致较长的通信延迟 和严重的通信竞争。针对这一问题,提出了一种基于闭合最小图划分模型的多作业分配优化方法。该方法以最小化通信延迟和消除通信竞争为出发点,通过建立闭合最小图划分模型,将多作业分配优化问题转化成闭合最小图划分问题,并设计闭合最小图划分算法来获得优化的多作业分配方案。
[1] http://www.top500.org [2] Geist A.Paving the Roadmap to EXASCALE [J].SciDAC Review,NUMBER 16Special Issue,2010 [3] Jose A P,Jose M A,Jose A L.Optimization-based MappingFramework for Parallel Applications [J].Journal of Parallel and Distributed Computing,2011,10(71):1377-1387 [4] Kumar V.Introduction to Parallel Computing (2nd Ed.) [M].Addison-Wesley Longman Publishing Co.,Inc.,Boston,MA,USA,2002 [5] Sinnen O,Sousa L A.Communication Contention in Task Sche-duling [J].IEEE Transactions on Parallel and Distributed Systems,2005,16(6):503-515 [6] http://www.nas.nasa.gov/publications/npb.html [7] 李秋阳.并行系统性能评估技术研究[D].天津:南开大学,2002 [8] Kim J,Dally W J,Scott S,et al.Technology-Driven,Highly-Scalable Dragonfly Topology [J].SIGARCH Comput.Archit.News,2008,36:77-88 [9] Moadeli M,Shahrabi A,Vanderbauwhede W,et al.Communication Modelling of the Spidergon NoC with Virtual Channels [C]∥Proceedings of the 2007International Conference on Parallel Processing.Washington,DC,USA,2007:76-76 [10] 杨晓东,陆松,牟胜梅.并行计算机体系结构—技术与分析[M].北京:科学出版社,2009 [11] Yang X J,Liao X K,Lu K,et al.The TianHe-1A Supercompu-ter:Its Hardware and Software [J].Journal of Computer Science and Technology,2011,26(3):344-351 [12] 王之元.并行计算可扩展性分析与优化:能耗、可靠性与计算性能[D].长沙:国防科学技术大学,2011 |
No related articles found! |
|