基于CA-PSO算法的云计算资源调度策略  

Resource Scheduling Strategy in Cloud Computing Based on Cultural Particle Swarm Optimization Algorithm

作　　者：孟令玺[1] 李洪亮[1] 

机构地区：[1]长春工业大学人文信息学院,吉林长春130122

出　　处：《计算机仿真》2013年第10期406-410,共5页Computer Simulation

摘　　要：为了提高云计算资源的利用率,保持负载平衡,提出一种基于文化粒子群算法的云计算资源调度算法。首先建立云计算资源调度的目标函数,然后采用文化粒子群算法对目标求解,并通过粒子群算法的主群体空间和文化算法的知识空间形成"双演化双促进"机制,提高全局搜索能力;最后采用仿真对CA-PSO性能进行测试。仿真结果表明,相对于GA、PSO算法,CA-PSO不仅克服标准粒子群算法的不足,同时获得更短的任务完成时间,使资源节点负载更加均衡,尤其对于大规模任务的云计算资源调度优势更加明显。In order to improve the utilization rate of cloud resource scheduling and keep load balance, this paper proposed a computing resource scheduling method based on cultural algorithm and particle swarm optimization algo- rithm. Firstly, the objective function of resource scheduling was established, and then the cultural particle swarm al- gorithm was used to solve the problem, in which the spatial knowledge main population space of particle swarm opti- mization algorithm and cultural algorithm formed the ＂dual evolution and dual promotion＂ mechanism to improve the global search capability and efficiency. Finally, the simulation test was carried out to test the performance of CA - PSO. The simulation results show that, compared with GA, PSO algorithm, the proposed method not only overcomes the shortcomings of the particle swarm algorithm, and shortens the task completion time. The load of resource node is more balance, especially for large - scale resource scheduling in cloud computing.

关 键 词：云计算 资源调度 文化算法 粒子群算法 

分 类 号：TP393[自动化与计算机技术—计算机应用技术]