摘要
面对云计算中的大量任务,为了对其进行高效的调度,缩短任务完成时间并提高资源利用率,对基于粒子群算法的云计算任务调度算法进行了研究.首先用自然数对粒子编码表示粒子的位置,并在解空间内随机初始化种群,每个粒子的位置对应一个可行的调度方案.每次迭代更新后,对粒子进行修复操作,为降低粒子跑出解空间的概率,同时对粒子速度进行限定.针对传统粒子群算法易陷入早熟的缺陷,加入混沌扰动策略使种群跳出局部最优.通过Cloudsim仿真平台进行实验测试,实验结果表明该算法能取得更好的调度结果并且收敛速度更快.
Facing with large amount of tasks in cloud computing, a task scheduling algorithm based on improved Particle Swarm Optimization(PSO) was taken into research to minimize the task completion time and maximize the resource utilization. Firstly, the velocity and position of each particle are set randomly within the search space in the initialization. A swarm of particles are used to represent the potential solutions and the position of each particle is encoded by natural number. After each iteration update, legalized method was used to repair the particle. In order to reduce the probability of particles running out of the solution space, some method was taken to limit the velocity of each particle. For overcoming precocious, chaos was combined with PSO. By using chaos disturbance, the particles can have a better position. Cloudsim was used to stimulate cloud computing environment for experimental test. Experimental results show that compared with tradition PSO the improved PSO converges faster and have a better scheduling result.
出处
《微电子学与计算机》
CSCD
北大核心
2016年第8期112-116,共5页
Microelectronics & Computer
基金
国家自然科学基金项目(61561001)
北方民族大学重点科研项目资助(2015KJ10)
北方民族大学研究生创新项目(YCX1562)