期刊文献+

一种基于参数扫描应用容错的粗粒度网格调度算法

A fault tolerant grid scheduling algorithm for coarse-grained tasks based on parameter sweep applications
下载PDF
导出
摘要 考虑作业处理时延以及作业传输时延对调度器性能的影响,在RR调度算法的基础上提出了一种新的基于参数扫描应用的调度算法PRR,并对该算法进行了理论分析,得出了该调度算法不需要作业以及处理器的相关信息,且具有比较好的容错性的结论.利用网格仿真软件Gridsim对PRR调度算法的理论分析结果进行了验证,并将该算法与RR调度算法的性能进行了比较.结果表明,当时延较大时,PRR调度算法性能改善比较明显. Considering the effect of delay in task submitting and data transferring on scheduler performance, a new scheduling algorithm PRR based on parameter sweep application was presented based on RR scheduling algorithm. The theoretical analysis shows that the new algorithm has high fault tolerance and does not need the related information of tasks and processors. With the simulation software Gridsim, the theoretical result of the PRR scheduling algorithm was verified, and the performance of the new scheduling algorithm at long delay was obviously improved as compared with that of the scheduling algorithm RR.
作者 张磊 王毅
出处 《河海大学学报(自然科学版)》 CAS CSCD 北大核心 2008年第2期258-262,共5页 Journal of Hohai University(Natural Sciences)
基金 国家自然科学基金(60403043) 河海大学“十五”211项目(GF1.3)
关键词 网格调度器 参数扫描 PRR调度算法 RR调度算法 容错调度器 grid scheduler parameter sweep PRR scheduling algorithm RR scheduling algorithm fault tolerant scheduler
  • 相关文献

参考文献10

  • 1CASANOVA H,LEGRAND A,ZAGORODNOV D,et al.Heuristics for scheduling parameter sweep applications in grid environments[C]//9th Heterogeneous Computing Workshop.Washington:IEEE Computer Society,2000:349-363.
  • 2PARANHOS D,CIRNE W,BRASILEIRO F.Trading cycles for information:using replication to schedule bag-of-tasks applications on computational grids[J].Lecture Notes in Computer Science,2003,2790:169-180.
  • 3BUCUR I D,EPEMA H J.Local versus global schedulers with processor co-allocation in multicluster systems[J].Lecture Notes in Computer Science,2002,2537:184-204.
  • 4MAHESWARAN M,ALI S,SIEGEL H J,et al.Dynamic matching and scheduling of a class of independent tasks onto heterogeneous computing systems[C]//8th IEEE Heterogeneous Computing Workshop.Washington:IEEE Computer Society,1999:30-44.
  • 5GRAHAM R L.Bounds for certain multiprocessing anomalies[J].Bell System Technical Journal,1996,45:1563-1581.
  • 6FUJIMOTO N,HAGIHARA K.Near-optimal dynamic task scheduling of precedence constrained coarse-grained tasks onto a computational grid[C]//The 2nd International Symposium on Parallel and Distributed Computing.Washington:IEEE Press,2003:80-87.
  • 7FUJIMOTO N,HAGIHARA K.A comparison among grid scheduling algorithms for independent coarse-grained tasks[C]//SAINT 2004 Workshop on High Performance Grid Computing and Networking.Washington:IEEE Press,2004:674-680.
  • 8MARTY H,JAREK G,JOE B,et al.State and events for web services:a comparison of five ws-resource framework and ws-notification implementations[C]//The 14th IEEE International Symposium on High Performance Distributed Computing.Washington:IEEE Computer Society,2005:24-27.
  • 9ALLEN A.Probability,statistics and queuing theory with computer science[M].Toronto:Application Academic Press,1978.
  • 10BUYYA R,MURSHED M.Gridsim:a toolkit for the modeling and simulation of distributed resource management and scheduling for grid computing[J].The Journal of Concurrency and Computation,2002,14:13-15.

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