Untitled

Abstract

Cloud Computing is a technique involving multiple resources being requested by versatile cloud users for allocation of shared resources. The tasks of application requests are allotted to virtual machines (VMs). In different situation different machines get different load. So, load balancing becomes necessary among different VMs. A decentralized load balancer is used to identify best number of tasks to be allocated to each VM. During load allocation to VM, for execution of tasks, the most optimal VM is identified for the load which is best capable of handling the task. The cloud user's application task is them mapped onto that VM to reduce the energy consumption and total execution time. In this paper, decentralized load balancing technique is used to distribute the load on each virtual machine which is enhanced using particle swarm optimization (PSO) which is a swarm based heuristic optimization technique. Moreover, the results are analyzed and compared with centralized load balancer for energy efficiency and throughput parameters.