计算机科学 ›› 2015, Vol. 42 ›› Issue (1): 33-37,46.doi: 10.11896/j.issn.1002-137X.2015.01.007

• 2013年全国理论计算机科学学术年会 • 上一篇    下一篇

基于随机Petri网的高性能计算系统作业调度及InfiniBand网络互连的性能分析

李智佳,胡翔,焦莉,王伟锋   

  1. 中国科学院软件研究所 北京100190;中国科学院大学计算机与控制学院 北京100049,中国科学院软件研究所 北京100190;中国科学院大学计算机与控制学院 北京100049,中国科学院软件研究所 北京100190,中国科学院软件研究所 北京100190;中国科学院大学计算机与控制学院 北京100049
  • 出版日期:2018-11-14 发布日期:2018-11-14
  • 基金资助:
    本文受国家科技重大专项(2102zx03039-004),中国科学院研究生科技创新专项资助

Performance Evaluation of Job Scheduling and InfiniBand Network Interconnection in High Performance Computing System Based on Stochastic Petri Nets

LI Zhi-jia, HU Xiang, JIAO Li and WANG Wei-feng   

  • Online:2018-11-14 Published:2018-11-14

摘要: 基于模型的分析技术在系统研究和设计中发挥着重要作用,它具有简单灵活、可扩展性强、高效等优点,其中随机Petri网在性能评价方面得到了广泛的应用。使用随机Petri网为高性能计算机的作业调度系统进行抽象和建模,并将其与InfiniBand网络互连结构相结合来整体分析用户作业的延迟等性能指标。实验表明,该方法是可行的,且具有相对较高的精度。

关键词: 随机Petri网,高性能计算,InfiniBand,作业调度,模型,性能评价

Abstract: Model-based analysis technology plays an important role in the system research and design,because it has many advantages,such as simplicity,flexibility,scalability,and efficiency.Especially,stochastic Petri nets are widely used in performance evaluation.We proposed a method of synthetical performance evaluation of job scheduling and InfiniBand network interconnection based on stochastic Petri nets.The experiment results demonstrate that our method is feasible and comparatively accurate.

Key words: SPN,HPC,InfiniBand,Job scheduling,Model,Performance evaluation

[1] Murata T.Petri nets:properties,analysis and applications[J].Proceedings of the IEEE,1989,7(4):541-580
[2] Petrini F,Kerbyson D J,Pakin S.The case of the missing supercomputer performance:achieving optimal performance on the 8,2 processors of ASCI Q[C]∥Proceedings of the 2003 ACM/IEEE Conference on Supercomputing,2003.Phoenix:ACM,2003:55
[3] Hu L,Gorton I.Performance evaluation for parallel systems:A Survey[R].Sydney:University of NSW,1997
[4] Ciardo G,Cherkasova L,Kotov V,et al.Modeling a scalablehigh-speed interconnect with stochastic Petri nets[C]∥Procee-dings of the Sixth International Workshop onPetri Nets and Performance Models,1995.Durham:IEEE Computer Society Press,1995:83-92
[5] Jain R.The art of computer systems performance analysis:techniques for experimental design,measurement,simulation,and modeling[M].New York:John Wiley & Sons,1991
[6] Brewer E A,Dellarocas C N,Colbrook A,et al.PROTEUS:a high-performance parallel-architecture simulator[C]∥Procee-dings of the 1992 ACM SIGMETRICS Joint International Conference on Measurement and Modeling of Computer Systems,1992.Newport:ACM,1992:247-248
[7] Florin G,Natkin S.Evaluation based upon stochastic petri nets of the maximum throughput of a full duplex protocol[C]∥Girault C,Reisig W,eds.Berlin:Springer-Verlag,1982:280-288
[8] Marsan M A,Conte G,Balbo G.A class of generalized stochastic Petrinets for the performance evaluation of multiprocessor systems[J].ACM Transactions of Computer Systems,1984,2(2):93
[9] Lin C,Marinescu D C.Stochastic high-level Petri nets and appli-cations[J].IEEE Transactions on Computers,1988,37(7):815
[10] Ciardo G,Muppala JK,Trivedi KS.SPNP:Stochastic Petri Net Package[C]∥Proceedings of 3rd International Workshop on Petri Nets and Performance Models,1989.Kyoto:IEEE Computer Society,1989:142-150 (下转第46页)(上接第37页)
[11] Zhou S N,Zheng X H,Wang J W,et al.UTOPIA:a load sharing facility for large,heterogeneous distributed computer systems[J].Software:Practiceand Experience,1993,23(12):1305-1336
[12] Mu’alem A W,Feitelson D G.Utilization,predictability,workloads,and user runtime estimates in scheduling the IBM SP2 with backfilling[J].IEEE Transactionson Parallel and Distributed systems,2001,12(6):529-543
[13] Sherwani J,Ali N,Lotia N,et al.Libra:a computational economy-based job scheduling system for clusters[J].Software:Practice and Experience,2004,34(6):573
[14] Shan Z G,Lin C.Modeling and performance evaluation of hierarchical job scheduling on the grids[C]∥Proceedings of the 2007 International Conference on Grid and Cooperative Computing,2007.Urumchi.IEEE Computer Society,2007:296-303
[15] Naik V K,Setia S K,Squillante M S.Performance analysis of job scheduling policies in parallel supercomputing environments[C]∥Proceedings of the 1993 ACM/IEEE conference on Supercomputing,1993.New York:ACM,1993:824-833
[16] Moschakis I A,Karatza H D.Evaluation of gang scheduling performanceand cost in a cloud computing system[J].The Journal of Supercomputing,2012,59(2):975
[17] Abawajy J H.An efficient adaptive scheduling policy for high-performance computing[J].Future Generation Computer Systems,2009,25(3):364
[18] Platform Computing Corporation.Running Jobs with PlatformLSF.http://www-03.ibm.com/systems/services/platformcomputing
[19] Palma J N.Performance evaluation of interconnection networks using simulation-tools and case studies[D].Bilbao:University of the Basque Country,2009
[20] Sur S,Koop M J,Chai L,et al.Performance analysis and evaluation of MellanoxConnectXInfiniBand architecture with multi-core platforms[C]∥Proceedings of the 15th Annual IEEE Symposium on High-Performance Interconnects,2007.Stanford:IEEE,2007:125-134
[21] Mellanox.MellanoxInfiniBand Training.http://www.me-llanox.com/
[22] 超级计算机TOP500排名.http://www.top500.org/
[23] 林闯.随机Petri网和系统性能评价(第2版)[M].北京:清华大学出版社,2005
[24] 陈永然.面向高性能计算的性能评价模型技术研究[D].长沙:国防科学技术大学,2007
[25] 王翠萍.LSF系统中作业调度的研究与优化[D].西安:西安电子科技大学,2009
[26] 曹宗雁.高性能计算集群运行时环境的配置优化[J].科研信息化技术与应用,2011,2(6):52-61
[27] 林闯.随机Petri网模型的精化设计[J].软件学报,2000,11(1):104-109

No related articles found!
Viewed
Full text


Abstract

Cited

  Shared   
  Discussed   
[1] 雷丽晖,王静. 可能性测度下的LTL模型检测并行化研究[J]. 计算机科学, 2018, 45(4): 71 -75, 88 .
[2] 夏庆勋,庄毅. 一种基于局部性原理的远程验证机制[J]. 计算机科学, 2018, 45(4): 148 -151, 162 .
[3] 厉柏伸,李领治,孙涌,朱艳琴. 基于伪梯度提升决策树的内网防御算法[J]. 计算机科学, 2018, 45(4): 157 -162 .
[4] 王欢,张云峰,张艳. 一种基于CFDs规则的修复序列快速判定方法[J]. 计算机科学, 2018, 45(3): 311 -316 .
[5] 孙启,金燕,何琨,徐凌轩. 用于求解混合车辆路径问题的混合进化算法[J]. 计算机科学, 2018, 45(4): 76 -82 .
[6] 张佳男,肖鸣宇. 带权混合支配问题的近似算法研究[J]. 计算机科学, 2018, 45(4): 83 -88 .
[7] 伍建辉,黄中祥,李武,吴健辉,彭鑫,张生. 城市道路建设时序决策的鲁棒优化[J]. 计算机科学, 2018, 45(4): 89 -93 .
[8] 刘琴. 计算机取证过程中基于约束的数据质量问题研究[J]. 计算机科学, 2018, 45(4): 169 -172 .
[9] 钟菲,杨斌. 基于主成分分析网络的车牌检测方法[J]. 计算机科学, 2018, 45(3): 268 -273 .
[10] 史雯隽,武继刚,罗裕春. 针对移动云计算任务迁移的快速高效调度算法[J]. 计算机科学, 2018, 45(4): 94 -99, 116 .