Pareto based ant lion optimizer for energy efficient scheduling in cloud environment

Abstract

The tremendous growth of cloud-based data centres leads to significant amount of energy. Thus, the prime concern for the cloud service providers is to generate environment friendly solution by minimizing the energy consumption. Further, due to increased demand of scientific workflow applications, cloud providers face a challenging issue of efficient workflow scheduling by minimizing the makespan. To address the above-mentioned contradictory issues, we proposed a Pareto based multi-objective discrete ant lion optimization algorithm (PBMO-DALO) to solve workflow-scheduling problem in cloud data centres with minimizing the conflicting objectives of makespan and energy consumption simultaneously. Distinguished from the original ant lion optimization algorithm, the proposed algorithm involves new encoding scheme for ants and antlions, their random walk and selection of fitter antlion for trap building to address the discrete nature of the workflow scheduling problem. Subsequently, the PBMO-DALO uses the Pareto dominance and crowding distance approach to tackle optimization of multiple objectives and to achieve the optimal solutions. The simulation results indicate that the proposed PBMO-DALO algorithm overwhelms other competing algorithms and generates good trade-off solutions with better convergence and uniform diversity.