作者投稿和查稿 主编审稿 专家审稿 编委审稿 远程编辑

计算机工程 ›› 2009, Vol. 35 ›› Issue (24): 96-98. doi: 10.3969/j.issn.1000-3428.2009.24.032

• 软件技术与数据库 • 上一篇    下一篇

基于QoS的网格负载调度算法

雷炳翰1,何 军1,何 翔1,李 奇2   

  1. (1. 四川大学计算机学院,成都 610065;2. 四川省计算机软件重点实验室,成都 610068)
  • 收稿日期:1900-01-01 修回日期:1900-01-01 出版日期:2009-12-20 发布日期:2009-12-20

Grid Load Schedule Algorithm Based on QoS

LEI Bing-han1, HE Jun1, HE Xiang1, LI Qi2   

  1. (1. College of Computer Science, Sichuan University, Chengdu 610065; 2. Key Lab of Software, Sichuan Province, Chengdu 610068)
  • Received:1900-01-01 Revised:1900-01-01 Online:2009-12-20 Published:2009-12-20

摘要: 针对多数启发式算法不能满足任务紧迫性需求的问题,基于Min-min算法和Sufferage算法的优点提出基于紧迫性需求的系统损失与任务复杂度平衡(QoS-based BSLTC)算法。在网格任务中加入QoS参数,在任务调度过程中,基于任务的分配损失和权衡系数,比较最大损失任务和最高计算复杂度任务,选出对系统性能影响最大的任务,并将该任务映射到完成时间最早的资源。分析结果表明,与Min-min算法和Sufferage算法相比,QoS-based BSLTC算法具有执行效率高、任务完成总时间短、负载均衡度高、兼顾任务紧迫性的优点。

关键词: 任务调度, QoS参数, 负载均衡度

Abstract: Aiming at the problem that most of the heuristic algorithm can not meet the task’s urgency demand, this paper presents the QoS-based Balance System Loss and Task Complexity(QoS-based BSLTC) algorithm based on the merits of Min-min algorithm and Sufferage algorithm. It joins the QoS parameter in the grid task. During the process of task schedule, the maximal loss task and the maximal computational complexity task is compared based on the distribution loss and measure coefficient. The task which has the most influence on the system is picked out and mapped to the resource with a minimum completion time. Analysis results verify that QoS-based BSLTC algorithm has merits of higher executing efficiency, needing less time to complete the task, high load balance degree and giving consideration to task urgency by comparing with Min-min algorithm and Sufferage algorithm.

Key words: task schedule, QoS parameter, load balance degree

中图分类号: