计算机工程与应用 ›› 2011, Vol. 47 ›› Issue (14): 62-65.

• 网络、通信、安全 • 上一篇    下一篇

融合小生境遗传禁忌的多目标网格调度研究

陈友文1,2,李智勇1,任长安1   

  1. 1.湖南大学 计算机与通信学院,长沙 410082
    2.湖南涉外经济学院 计算机科学与技术学部,长沙 410205
  • 收稿日期:1900-01-01 修回日期:1900-01-01 出版日期:2011-05-11 发布日期:2011-05-11

Multi-objective grid task scheduling algorithms based on combined strategy of Niche genetic algorithm and Tabu search

CHEN Youwen1,2,LI Zhiyong1,REN Chang’an1   

  1. 1.Department of Computer Science and Communication,Hunan University,Changsha 410082,China
    2.Department of Computer,Hunan International Economics University,Changsha 410205,China
  • Received:1900-01-01 Revised:1900-01-01 Online:2011-05-11 Published:2011-05-11

摘要: 针对在动态、复杂的网格系统中,资源的失效非常频繁,影响网格计算的服务质量和效率的问题,提出了基于融合小生境的遗传禁忌的多目标网格任务调度算法,将任务生存性和任务完成时间(Makespan)结合起来,给出一个可调节的多目标集成效用函数,实现了基于融合小生境的遗传禁忌的多目标启发式调度算法。模拟实验结果表明,该算法能够有效地平衡任务生存性和Makespan两个目标,更适合开放复杂的网格计算环境。

关键词: 遗传算法, 禁忌搜索算法, 融合, 多目标, 网格, 任务调度

Abstract: Combining the advantage of genetic algorithm and tabu search,a combined evolutionary computing is proposed to optimize the grid task scheduling strategy.In the complex and dynamic grid system,failure of resources is very common.To mitigate this problem,a local objective function which can be adjusted is proposed by incorporating the makespan and the survivability of task.The simulation results show that the scheduling algorithm can trade off these two objectives.So it can be applied in the complex grid computing environment well.

Key words: genetic algorithm, tabu search, combined, multi-objective, grid, task scheduling