An Optimal Bi-Objective Particle Swarm Optimization Algorithm for Task Scheduling in Cloud Computing

Abstract

Cloud Computing is an innovative technology that has brought a revolutionary transformation in the way computing services such as processing, storage and networks are delivered to the users. Quality of Service (QoS) greatly depends on how well the services are scheduled on the available resources. In previous approaches, several combination of QoS parameter(s) are considered to optimize task scheduling algorithms. They focused on parameters such as execution cost, budget, deadline, and execution time. This paper proposes an efficient task scheduling that explicitly considers reliability and execution time to optimize task scheduling in cloud computing environment. For this purpose, the concept of Vector Evaluated Particle Swarm optimization (VEPSO) is utilized. Two Swarms are used one for each objective such that the information of one swarm is used to update the velocities of the other swarm. This way, both the swarms cooperate together to get a better set of solution. The results when compared with the basic PSO using weighted sum approach are found to be better in terms of execution time, throughput and failure rate.