Efficient resource allocation in cloud data centers through genetic algorithm

Abstract

Efficient resource scheduling and managing available resources is a great concern in modern data centers. Cloud computing which is based on virtualization automates the process of distributing available resources among virtual machines. This paper proposes an enhanced version of resource allocation algorithm based on genetic algorithm. In other words, the decision accuracy in the process of virtual machine allocation to proper physical machines is enhanced. Usually, CPU and RAM of physical machines are considered in decision algorithms. However, this paper shows that to have a balanced cloud, it is necessary to consider more parameters in resource allocation algorithm. Moreover, the importance of considering weights for the parameters in decision algorithm is presented. It is shown that some parameters are more important than the others and should have higher impact in fitness function of genetic algorithm.