An Enhanced Task Scheduling Algorithm on Cloud Computing Environment

Abstract

Cloud computing is the technology that moves the information technology (IT) services out of the office.Unfortunately, Cloud computing has faced some challenges.The task scheduling problem is considered one of the main challenges because a good mapping between the available resources and the users' tasks is needed to reduce the execution time of the users' tasks (i.e., reduce make-span), and increase resource utilization.The objective of this paper is to introduce and implement an enhanced task scheduling algorithm to assign the users' tasks to multiple computing resources.The aim of the proposed algorithm is to reduce the execution time, and cost, as well as, increase resource utilization.The proposed algorithm is considered an amalgamation of the Particle Swarm Optimization (PSO),the Best-Fit (BF), and Tabu-Search (TS) algorithms; called BFPSOTS.According to the proposed BFPSOTS algorithm, the BF algorithm has been used to generate the initial population of the standard PSO algorithm instead of to be random.The Tabu-Search (TS) algorithm has been used to improve the local research by avoiding the trap of the local optimality which could be occurred using the standard PSO algorithm.The proposed hybrid algorithm (i.e., BFPSOTS) has been implemented using Cloudsim.A comparative study has been done to evaluate the performance of the proposed algorithm relative to the standard PSO algorithm using five problems with different number of independent task, and Virtual Machines (VMs).The performance parameters which have been considered are the execution time (Makspan), cost, and resources utilization.The implementation results prove that the proposed hybrid algorithm (i.e., BFPSOTS) outperforms the standard PSO algorithm..