计算机应用 ›› 2010, Vol. 30 ›› Issue (05): 1316-1320.

• 软件过程技术 • 上一篇    下一篇

异构系统中的综合性启发式任务调度算法

赵欢1,江文2,李学辉2   

  1. 1.
    2. 湖南大学
  • 收稿日期:2009-11-02 修回日期:2010-01-28 发布日期:2010-05-04 出版日期:2010-05-01
  • 通讯作者: 江文
  • 基金资助:
    湖南省科技计划项目

Synthesized heuristic task scheduling algorithm for heterogeneous system

  • Received:2009-11-02 Revised:2010-01-28 Online:2010-05-04 Published:2010-05-01
  • Contact: Jiang Wen

摘要: 任务的单个属性常作为基于优先驱动的表调度算法的优先级,针对这种方法常出现优先级相同的情况,提出一个综合性启发式算法HCPFS。算法分三个优先级选择任务进行调度,从高到低依次为:关键路径上的任务、就绪任务到出口任务的路径长度和后继任务数。调度过程中,算法采用任务复制和空闲时间区段任务插入的方法。采用随机生成图法和任务图集进行了算法模拟和比较,实验数据表明HCPFS算法具有更好的调度性能。

关键词: 异构计算系统, 综合性启发式算法, 关键路径, 任务复制

Abstract: The list-scheduling algorithm driven by priority chooses tasks to schedule only by one attribute. But in the case of that different tasks should have the same priority, it can not work well. Therefore, a new synthesized heuristic algorithm called HCPFS was proposed. There were three levels of priority in the algorithm to choose task. First, the tasks in critical path have the highest priority. Secondly, the tasks with longer path to exit task should be selected, and then the tasks with more successors were chosen to schedule. The algorithm took duplication method and insertion method in the process of scheduling task. The experimental results in randomly generated graphs and sets of task graphs show that HCPFS performs well compared with HEFT and HCNF in heterogeneous environment.

Key words: heterogeneous computing system, synthesized heuristic algorithm, critical path, task duplication