Cloud Computing Task Scheduling Based On Improved Differential Evolution Algorithm

With the application and development of big data,countless experts and scholars have studied the hot technology of cloud computing.This technology has powerful data processing capability.Therefore,in the face of massive data resources,how to carry out reasonable scheduling,so that the utilization of resources and the efficiency of users' task execution can reach the optimal degree,is one of the key issues studied by cloud computing technology.This paper focuses on the use of improved differential evolution algorithm in cloud computing task scheduling problems,based on the traditional differential evolution(DE)algorithm.This algorithm is a global optimization algorithm emerged in recent years,which has the advantages of being easy to implement,easy to use,fast and reliable,but it also has the problems of slow convergence,easy premature maturity and difficult parameter setting.To address these problems,three improved differential evolution cloud task scheduling algorithms are proposed in this paper,namely Piecewise Differential Evolution Algorithm(PDE),Adaptive Mutation Differential Evolution Algorithm(AMDE)and Piecewise Adaptive Mutation Differential Evolution(PAMDE).Evolution Algorithm(AMDE)and Piecewise Adaptive Mutation Differential Evolution(PAMDE).The maximum task completion time is optimized by improving the scaling factor F,crossover factor CR and variation strategy.Simulation experiments are conducted to demonstrate that the maximum task completion time of the three improved differential evolution cloud task scheduling algorithms is less than that of the traditional differential evolution algorithm,and the performance optimization of the algorithms becomes more obvious the larger the number of tasks.