计算机应用 ›› 2012, Vol. 32 ›› Issue (05): 1418-1420.

• 先进计算 • 上一篇    下一篇

基于蚁群优化算法的云计算任务分配

张春艳1,刘清林1,孟珂2   

  1. 1. 中国矿业大学 计算机科学与技术学院,江苏 徐州 221116
    2. 中国矿业大学 计算机科学与技术学院,江苏 徐州 221008
  • 收稿日期:2011-11-16 修回日期:2012-01-16 发布日期:2012-05-01 出版日期:2012-05-01
  • 通讯作者: 张春艳
  • 作者简介:张春艳(1985-),女,江苏沛县人,硕士研究生,主要研究方向:蚁群算法、云计算;
    刘清林(1985-),男,山东泰安人,硕士研究生,主要研究方向:协同推荐系统;
    孟珂(1987-),男,江苏沛县人,硕士研究生,CCF会员,主要研究方向:最短路径算法。

Task allocation based on ant colony optimization in cloud computing

ZHANG Chun-yan1,LIU Qing-lin1,MENG Ke2   

  1. 1. School of Computer Science and Technology, China University of Mining and Technology, Xuzhou Jiangsu 221116, China
    2. School of Computer Science and Technology, China University of Mining and Technology, Xuzhou Jiangsu 221008, China
  • Received:2011-11-16 Revised:2012-01-16 Online:2012-05-01 Published:2012-05-01
  • Contact: ZHANG Chun-yan

摘要: 针对已有的适用于分配任务的蚁群算法易陷入局部最优解的缺陷,提出了一个保证云服务质量的分组多态蚁群算法。该算法将蚁群按职能不同分为搜索蚁、侦察蚁和工蚁,根据预测完成时间的更新使平均完成时间逐渐取得最小值,从而减少产生局部最优解的可能,最后通过Cloudsim仿真实现。结果表明该方法减少了处理请求任务的平均完成时间,提高了任务处理的效率。

关键词: 云计算, Cloudsim, 蚁群算法, 多态

Abstract: Concerning the defects of the Ant Colony Optimization (ACO) for the task allocation, a grouping and polymorphic ACO was proposed to improve the service quality. The algorithm, which divided the ants into three groups: searching ants, scouting ants and working ants, with the update of forecast completion time to gradually get the minimum of the average completion time and to decrease the possibility of generation to local optimum, was emulated and achieved with Cloudsim tookit at last. Results of the experiment show that the time of handling requests and tasks of this approach has been reduced and the efficiency of handling tasks gets improved.

Key words: cloud computing, Cloudsim, Ant Colony Optimization (ACO), polymorphic

中图分类号: