A Metaheuristic Algorithm for Job Scheduling in Grid Computing

Abstract

These days the number of issues that we can not do on time is increasing. In the mean time, scientists are trying to make questions simpler and using computers. Still, more problems that are complicated need more complex calculations by using highly advanced technology. Grid computing integrates distributed resources to solve complex scientific, industrial, and commercial problems. In order to achieve this goal, an efficient scheduling system as a vital part of the grid is required. In this paper, we introduce CUckoo-Genetic Algorithm (CUGA), which inspired from cuckoo optimization algorithm (COA) with genetic algorithm (GA) for job scheduling in grids. CUGA can be applied to minimize the completion time of machines, and it could avoid trapping in a local minimum effectively. The results illustrate that the proposed algorithm, in comparison with GA, COA, and Particle Swarm Optimization (PSO) is more efficient and provides higher performance. Index Terms—Cuckoo optimization algorithm, genetic algorithm, job scheduling, grid computing.