Optimized Genetic Algorithm for Efficient Allocation of Virtualized Resource in Cloud (OGA_EAVRC)

Abstract

Efficient allocation of achievable virtual resources to the diverse users is a key challenging issue in a controlled and collaborative cloud environment. As well, balancing the load among the resources and mapping these virtual resources to the physical machines is even bigger challenge of present distributed computing arena. Numerous approaches were introduced by different researchers including Genetic Algorithm for dealing with these challenges, but their scope was limited to certain specific performance elements. Hence, there is a need of optimizing the existing research implementations for efficient allocation of virtualized resources in cloud computing environment. Usually, in a typical distributed computing environment like cloud computing, allocation of virtual resource and balancing of workload among them is realized by means of virtual machines live migration. This article introduces an optimization of existing Genetic algorithm (GA) that mainly intended for VM resource provisioning and load balancing. The proposed OGA_EAVRC considers Population size, Fitness function, Mutation probability, and success rate of resource for optimizing the performance through efficient resource allocation. Key objective of this work is to utilize each physical resource effectively and allocated them to end users efficiently. For studying the operational performance of OGA_EAVRC, an event based CloudSim was chosen. Simulation results states that the proposed OGA_EAVRC can efficiently allocates the workload among virtualized resource by reducing VM's migration among the physical machines.