计算机应用 ›› 2014, Vol. 34 ›› Issue (2): 357-359.

• 先进计算 • 上一篇    下一篇

云环境下基于改进遗传算法的虚拟机调度策略

袁爱平1,万灿军2   

  1. 1. 长沙民政职业技术学院 软件学院,长沙 410004;
    2. 湖南大学 计算机与通信学院,长沙 410082
  • 收稿日期:2013-08-21 修回日期:2013-10-17 出版日期:2014-02-01 发布日期:2014-03-01
  • 通讯作者: 袁爱平
  • 作者简介:袁爱平(1975-),男,湖南邵阳人,高级工程师,硕士,主要研究方向:嵌入式系统、计算机网络、知识工程;万灿军(1982-),男,湖南岳阳人,博士研究生,主要研究方向:可信软件、软件动态演化、高效能计算。
  • 基金资助:
    湖南省教育规划基金项目;湖南省科技计划项目

Virtual machine deployment strategy based on improved genetic algorithm in cloud computing environment

YUAN Aiping1,WAN Chanjun2   

  1. 1. Department of Software, Changsha Social Work College, Changsha Hunan 410004, China;
    2. School of Computer and Communication, Hunan University, Changsha Hunan 410082, China
  • Received:2013-08-21 Revised:2013-10-17 Online:2014-02-01 Published:2014-03-01
  • Contact: YUAN Aiping

摘要: 针对云环境下服务器内部多种资源间分配不均衡问题,提出了一种多维资源协同聚合的虚拟机调度算法MCCA。该算法在分组遗传算法的基础上,采用模糊逻辑及基于资源利用率多维方差的控制参量,设计适应度函数指导搜索解空间。算法使用基于轮盘赌法的选择方法,并对交叉和变异等进行了优化,以实现快速有效地获取近似最优解。在CloudSim环境下进行了仿真,实验结果表明该算法对均衡多维资源分配和提高资源综合利用率具有一定的优势。

关键词: 云计算, 虚拟机, 多维均衡, 分组遗传算法

Abstract: Aiming at improving the resource utilization of data center by balanced usage of multiple resources, a scheduling algorithm based on group genetic algorithm for multi-dimensional resources coordination was proposed to solve the virtual machine deployment problem. To guide the solution searching, a fuzzy logic based multi-dimensional fitness function was raised. Meanwhile, innovative optimization of crossover and mutation was put forward to improve the solution quality. The results of simulation in CloudSim environment prove that using the proposed algorithm can obtain better multi-dimensional resources performance and higher resource utilization rate.

Key words: cloud computing, virtual machine, multi-dimensional balancing, group genetic algorithm

中图分类号: