云计算环境下基于蚁群优化的任务负载均衡调度算法  

Load balancing of task scheduling based on ant colony optimization in cloud computing environment

作　　者：赵梦[1] 李蜀瑜[1] 

机构地区：[1]陕西师范大学计算机科学学院,陕西西安710119

出　　处：《电子设计工程》2016年第8期30-33,共4页Electronic Design Engineering

基　　金：国家自然科学基金面上项目(41271387)

摘　　要：随着云计算的蓬勃发展,针对云计算中虚拟机负载不均衡及任务集完成时间较长的问题,提出了一种基于蚁群优化的任务负载均衡调度算法(WLB-ACO)。首先基于当前虚拟机的资源利用情况判断虚拟机的负载状态,其次,根据虚拟机的负载因子定义信息素的挥发因子(w),改进信息素更新规则,并利用WLB-ACO合理的分配任务,使整个系统达到负载均衡状态的同时任务集的完成时间最短。最后,采用Cloudsim工具设计仿真实验,实验结果表明,提出的基于蚁群优化的任务调度算法在性能、最短完成时间及算法的稳定收敛性上有了一定的提高。With the rapid development of cloud computing, cloud computing for the virtual machine load imbalance and take a long time to complete tasks set, it is proposed based on ant colony optimization task scheduling algorithm load balancing（WLB-ACO）. Firstly, the current virtual machine based on the resource utilization in order to determine a load state of the virtual machine, and secondly, the pheromone volatile factors（w） is defined by virtual machine ＇s load factor. To improve the pheromone update rules, and use WLB-ACO rational allocation of tasks, completion of the entire system to achieve load balancing state while the minimum set of tasks. Finally, the use of simulation tools to design Cloudsim experimental results show that the task scheduling based on ant colony optimization algorithm in terms of performance, the shortest time to complete a certain increase stability and convergence of the algorithm.

关 键 词：云计算 任务调度 负载均衡 最短完成时间 蚁群优化 

分 类 号：TN915[电子电信—通信与信息系统]