CLOUD COMPUTING RESOURCE SCHEDULING METHOD BASED ON DISTRIBUTION CHARACTERISTIC LEARNING GREY WOLF OPTIMIZATION

Aimed at the problems of low accuracy and falling into local optima easily, a cloud computing resource scheduling method based on improved grey wolf optimization is proposed. In the framework mode based on Map/Reduce, cloud computing resource scheduling mathematical model was constructed. In order to improve the global search ability of grey wolf optimization algorithm, the distribution characteristic learning framework which adjusted the search direction of algorithm was adopted. The improved grey wolf optimization was used to solve the cloud computing resource scheduling problem. The simulation experiment results show that compared with other algorithms, the improved grey wolf optimization has less convergence accuracy in solving resource scheduling problem, and can optimize a better resource scheduling strategy, especially in large-scale tasks condition.