Workflow Scheduling in Cloud Computing Using Memetic Algorithm

Abstract

Cloud computing is a distributed computing system which can be used to efficiently solve large size computational problems. On of the main models used cloud computing is the ”pay as you go” model for the provided service. Workflow can be scheduled on virtual machines in data centers to use resources efficiently. Workflow scheduling known as one of the NP-complete problems, due to the dynamism and heterogeneity nature of cloud systems. In this paper, a Memetic Algorithm (MA) is proposed to solve the cloud workflow-scheduling problem considering cost and deadline as two objectives to optimize scheduling of scientific workflow in a cloud-computing environment. The proposed algorithm used hill climbing local search algorithm as an extra operator for Genetic Algorithm (GA) to improve individual solutions during global search. The experiment results show that the proposed MA decreases the makespan of the workflow and performs better compared to GA and Particle Swarm Optimization (PSO) algorithms.