期刊文献+

可扩展并行作业调度模拟器ParaSim设计与应用 被引量:1

Design and Application of Scalable Parallel Job Scheduling Simulator ParaSim
下载PDF
导出
摘要 针对作业调度研究的需求,设计和实现了一个可扩展的并行作业调度模拟器ParaSim。ParaSim采用与实际并行作业系统近似的工作流程,以资源占用矩阵来表示计算资源,使用事件驱动的模式进行模拟调度和运行,支持空间共享和时间共享等多种调度策略,并允许对各调度参数进行设置。ParaSim已投入实际使用,为并行机作业调度策略的定量分析、调整和优化提供了有力的支持。 This paper introduces a scalable parallel job simulator named ParaSim.ParaSim makes use of the event-driven executing model and adopts matrix to represent computation resources,so it can support both space-sharing and time-sharing scheduling strategies in the same framework.It also provides many policies and scheduling parameters like real system.At present,ParaSim has been implemented and used in scheduling strategy research.The experiment based on ParaSim shows that it can benefit from ParaSim on research and selection of scheduling strategies in real job system.
出处 《计算机工程》 CAS CSCD 北大核心 2007年第18期45-47,共3页 Computer Engineering
基金 中国工程物理研究院基金资助项目(20060646)
关键词 大规模并行计算机 作业调度 调度评价 模拟器 massive parallel computer job scheduling evaluation of scheduling simulator
  • 相关文献

参考文献5

  • 1Feitelson D G.The Forgotten Factor:Facts on Performance Evaluation and Its Dependence on Workloads[C]//Proc.of Euro-Par 2002 Parallel Processing.2002:49-60.
  • 2Feitelson D G,Rudolph L.Metrics and Benchmarking for Parallel Job Scheduling[C]//Proc.of the 4th Workshop on Job Scheduling Strategies for Parallel Processing,Orlando.1998:1-24.
  • 3Franke H,Jann J,Moreira J.An Evaluation of Parallel Job Scheduling for ASCI Blue-pacic[C]//Proceedings of SC'99,Portland,Oregon.1999:11-18.
  • 4Krevat E,Castanos J,Moreira J.Job Scheduling for the Blue Gene/L System[C]//Proc.of the 8th Workshop on Job Scheduling Strategies for Parallel Processing,Edinburgh.2002:38-54.
  • 5Kleban S D,Clearwate S H.Simulating Performance Sensitivity of Supercomputer Job Parameters[C]//Proc.of High Performance Computing Symposium.2003.

同被引文献7

  • 1蒋伟进,王璞.基于MAS市场机制的动态计算资源调度模型研究[J].计算机研究与发展,2007,44(1):29-36. 被引量:9
  • 2Keller A,Reinefeld A.CCS Resource management innetworked HPC systems. Proc of HeterogenousComputing Workshop . 1998
  • 3Wolski R,Plank J S,Bryan T,et al.G-commerce:market formulations controlling resource allocation onthe computional grid. Proc of Int′l Parallel andDistributed Processing Symp . 2001
  • 4Lee C B,Snavely A E.Precise and realistic utilityfunctions for user-centric performance analysis ofscheduler. Proc of 16th IEEE Int’’l Symp HighPerformance Distrbuted Computing . 2007
  • 5Feitelson D.Parallel workloads archive. http:∥www.cs.huji.ac.il/labs/par-allel/workload/ . 2010
  • 6Buyya R,Abramson D,Giddy J.Nimrod/G: An Architecture for a Resource Management and Scheduling System in a Global Computational Grid. Proceedings of the Fourth International Conference on High Performance computing in Asia-Pacific Region . 2000
  • 7刘辉,龚斌,陈晓红,武蕾.基于激励与约束机制的网格资源分配模型[J].华中科技大学学报(自然科学版),2007,35(S2):20-23. 被引量:1

引证文献1

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