多目标遗传算法在云计算任务调度中的应用

Application of multiple-objective genetic algorithm for task scheduling in cloud environment

执行时间、执行成本和负载均衡是云环境中的主要优化目标,针对云计算环境中的任务调度问题,提出一种改进的多目标遗传算法。算法对目标进行了规范化,改进了加权求和的过程,并引入基于排列的选择方案和"最优比较法"的变异方案。最后在两个云计算场景中进行实验,对实验结果进行分析和统计,验证了算法的有效性和可行性。

Executing time,executing cost and load balancing are main optimization goals in cloud computing. For task scheduling in cloud computing environment,this paper proposes an improved multiobjective genetic algorithm. The algorithm improves weighted sum with normalized objective value, introduces selection of sorting and the variation of best scheme comparison. Finally it applied the algorithm to the simulation experiments in two cloud computing scene,the statistical experimental results verify the effectiveness and availability of the cloud scheduling method which is proposed by this thesis.