Service cost of resource scheduling in cloud computing based on an improved algorithm combining support vector machine with genetic algorithm

Abstract

It is important to achieve high effective scheduling in cloud computing environment and further guarantee quality of cloud service. In order to do it, fitness function is designed to synthetically reflect completion time and cost of tasks. Based on cross over operation of genetic algorithm and components selection of partial regression, some significant parameters in classification algorithm of support vector machine (SVM)are redefined to strengthen mutual learning capability and sorting ability between each items of subgroup. Meanwhile, it can improve convergence performance of algorithm. Moreover, from the view of results in simulation experiment, the proposed method represents better convergence performance and resource scheduling capability in the different number of resources than artificial neural network (ANN) and genetic algorithm (GA). In the final, according to the proposed method, high effective scheduling will be acquired and service cost may be reduced in cloud computing environment.