Energy efficient load balancing aware task scheduling in cloud computing using multi-objective chaotic darwinian chicken swarm optimization

Abstract

Scheduling of tasks in a cloud environment has larger influence on time and energy depletion. Different heuristic models were developed to solve the NP-hard task scheduling problem based on time. However, ideal task scheduling algorithms must also maximize energy efficiency with good load balancing and ensure better Quality-of-Service (QoS). An innovative multi-objective Chaotic Darwinian Chicken Swarm Optimization (CDCSO) system is suggested in this article to provide energy efficient QoS and load balancing aware task scheduling. The multi-objective CDCSO algorithm incorporates the chaotic and Darwinian Theory to the standard Chicken Swarm Optimization to increase its global exploration and maximize the convergence rate. This performance enhanced CDCSO algorithm models the cloud task scheduling problem as NP-hard and utilizes the optimization principles to solve them based on multiple objective parameters. The multi-objective fitness function used in CDCSO is modelled based on the objective parameters namely energy, cost, task completion time, response time, throughput and load balancing index. Based on this multi-objective function, the CDCSO effectively allocates the tasks to the suitable energy efficient, cost and time minimized Virtual machines (VMs) which are also optimally load balanced. CloudSim simulations were conducted and the obtained results illustrated that the proposed multi-objective CDCSO has provided better task scheduling with minimized energy, cost, time and optimal load balancing.