An Improved Differential Evolution Task Scheduling Algorithm Based on Cloud Computing

Abstract

It is a key issue to handle many tasks efficiently in cloud computing at low cost. For the cloud computing scheduling problem, in order to efficiently and reasonably assign a large number of tasks submitted by users to cloud computing resources, a task scheduling algorithm (IDE) based on improved differential evolution is proposed to consider both task completion time and cost dual objectives. The algorithm introduces an immune operator into the traditional differential evolution algorithm. According to the vaccination probability, the population is vaccinated during the iterative process to speed up the convergence of the algorithm. Introducing the judgment mechanism on the selection strategy can shorten the running time of the algorithm and effectively improve the shortcomings of the standard differential evolution algorithm with slow convergence speed. The original fixed scaling factor F becomes adaptive, which helps to increase the diversity of the population. The simulation experiment of the proposed algorithm is performed on the cloud computing platform CloudSim. Comparing the IDE algorithm with the traditional differential evolution algorithm, genetic algorithm and Min-Min algorithm, the results show that IDE algorithm task completion time is short, which improves the utilization of cloud computing resource pools, and the cost of computing resources in a similar period of time is low.