期刊文献+

基于资源监控的网格调度系统

Grid scheduling system based on resource monitoring
下载PDF
导出
摘要 研究ChinaGrid生物信息学网格中,根据资源负载变化进行动态作业调度的方案.针对中国网格公共支撑平台(CGSP)的执行管理模块在调度作业时,没有将计算节点的动态性作为调度优化的考虑因素,在China-Grid生物信息学网格中实现了一个基于资源监控的作业调度系统(SSRM).SSRM利用CGSV来实时获取资源的动态信息,并报告给ChinaGrid生物信息学网格的调度模块进行调度方案的优化.实验表明,与资源监控相结合的网格调度系统可以明显地缩短一组作业的完成时间. The execution management in ChinaGrid supporting platform addresses the problem of job scheduling without considering the dynamic status of resources.In this paper,we implement a scheduling system based on resource monitoring(SSRM) in bioinformatics platform of ChinaGrid.In SSRM,resources status is monitored by CGSV(ChinaGrid super vision) and reported to the scheduling module.Scheduling module optimizes the scheduling scheme according to the history measurements of jobs and current status of resources.The experimental result shows that the CTT(completion time of task) of a set of jobs in SSRM can be improved remarkably.
出处 《华中科技大学学报(自然科学版)》 EI CAS CSCD 北大核心 2007年第S2期76-78,共3页 Journal of Huazhong University of Science and Technology(Natural Science Edition)
基金 国家自然科学基金资助项目(90412006 90412011 60573110 90612016 60673152) 国家重点基础研究发展计划资助项目(2004CB318000 2003CB317007) 国家高技术研究发展计划资助项目(2006AA01A101 2006AA01A108 2006AA01A111)
关键词 网格计算 调度 资源管理 资源监控 基于资源监控的作业调度系统 grid computing scheduling resource management resource monitoring scheduling system based on resource monitoring(SSRM)
  • 相关文献

参考文献9

  • 1James Frey,Todd Tannenbaum,Miron Livny,Ian Foster,Steven Tuecke.Condor-G: A Computation Management Agent for Multi-Institutional Grids[J]. Cluster Computing . 2002 (3)
  • 2Foster I,Kesselman C,Tuecke S.The anatomyofthe grid:enabling scalable virtual organizations. Interational Journal of High Performance ComputingApplications . 2001
  • 3Jin H.ChinaGrid:making grid computing a reality. Digital Libraries:International Collaborationand Cross-Fertilization,Proceedings Lecture Notes inComputer Science . 2004
  • 4Wu Y W,Wu S,Yu H S,et al.Introduction to Chi-naGrid support platform. Lectuer Notes in Com-puter Science . 2005
  • 5Zheng W M,Liu L,Hu M Z,et al.CGSV*:an a-daptable stream-integrated grid monitoring system. Network and Parallel Computing,ProceedingsLecture Notes in Computer Science . 2005
  • 6Ma Ruyue,Meng Xiangxu,Liu Shijun,et al.Execu-tion management in ChinaGrid supporting platform. The 5th International Conference on Grid andCooperative Computing(GCC 2006) . 2006
  • 7Frey J,Tannenbaum T,Foster I,et al.Condor-G:acomputation management agent for multi-institutionalgrids. Journal of Cluster Computing . 2002
  • 8Berman F,Wolski R,Figueira S,et al.Application-level scheduling on distributed heterogeneous net-works. Proc Supercomputing′96 . 1996
  • 9Cooper K,,Dasgupta A,Kennedy K,et al.New Gridscheduling and rescheduling methods in the GrADSproject. Parallel and Distributed ProcessingSymposium . 2004

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