计算机工程与应用 ›› 2010, Vol. 46 ›› Issue (24): 1-3.DOI: 10.3778/j.issn.1002-8331.2010.24.001

• 博士论坛 • 上一篇    下一篇

结合价值密度和截止期的网格实时作业调度

李荣胜1,赵文峰2,徐惠民1   

  1. 1.北京邮电大学 信息与通信工程学院,北京 100876
    2.北京邮电大学 网络与交换技术国家重点实验室,北京 100876
  • 收稿日期:2010-04-07 修回日期:2010-07-05 出版日期:2010-08-21 发布日期:2010-08-21
  • 通讯作者: 李荣胜

Scheduling real-time grid jobs based on value density and relative deadline

LI Rong-sheng1,ZHAO Wen-feng2,XU Hui-min1   

  1. 1.School of Information and Communication Engineering,Beijing University of Posts and Telecommunications,Beijing 100876,China
    2.State Key Laboratory of Networking and Switching Technology,Beijing University of Posts and Telecommunications,Beijing 100876,China
  • Received:2010-04-07 Revised:2010-07-05 Online:2010-08-21 Published:2010-08-21
  • Contact: LI Rong-sheng

摘要: 在商业网格和云计算环境中,作业有到达时间、计算量、预算、截止期等参数,其中,预算是时间的函数。准确区分作业的重要性和紧迫性是作业调度系统的一个关键问题。综合利用这四个参数来定义作业的优先级,并提出基于价值密度和相对截止期的网格作业调度算法。分别对弱实时和强实时网格作业的调度进行仿真。仿真结果显示,所提出的调度算法的性能在两种情况下都优于所有对比算法的性能,且在强实时作业情况下优势更明显。

关键词: 价值密度, 优先级, 作业调度, 网格计算, 实时作业

Abstract: Jobs have arrival time,workload,budget and deadline parameters in economy-based grid computing and cloud computing environment.And the budget parameter changes over time.It is vital to differentiate jobs’ importance and urgency exactly for job scheduling system.These four parameters are all taken into account in defining a new priority of jobs here.Based on this priority,a value density and relative deadline jointly based scheduling algorithm is proposed.Results of simulation show that the proposed scheduling algorithm outperforms the referenced algorithms in scheduling of soft and hard real-
time grid jobs,especially in the latter.

Key words: value density, priority, job scheduling, grid computing, real-time job

中图分类号: