云环境下基于改进PSO算法的任务调度策略

Task Scheduling in Cloud Environment Based on Improved PSO Algorithm

摘要: 在云环境中，如何高效的实现任务与资源之间的映射一直是研究的热点问题。仿生算法被广泛的应用于任务调度的优化，其中粒子群优化算法凭借其结构简单、参数少和易实现的优点而受到大力重视。但是，标准的粒子群优化算法对于存在较多局部极值云计算环境，很容易出现“早熟”的问题，这是由于种群多样性的匮乏所致。因此，在搜寻过程中能够保持种群的多样性十分重要，本文提出了“分裂”的思想，经过实验证明，该方法能够很好的解决“早熟”的问题，有效的保持种群多样性，并能有效的减少任务的完成时间。

Abstract: In a cloud environment, how to map between tasks and resources efficiently has been a hot research topic. Bionic algorithm is widely used in task scheduling optimization, particle swarm optimization algorithm has been paid more attention by virtue of its simple structure, less parameters and easy to achieve. Standard particle swarm optimization algorithm, however, there are more local extreme in cloud computing environment, is prone to make the problem of “premature”, which is due to the lack of population diversity. Therefore, it is very important to maintain the diversity of the population in the search process, the paper put forward the idea of a “separatist”, the experiments show that the method can solve the problem of “premature” well, it is effective to maintain the diversity of the population, and reduce the completion time of the task effectively.