A Low-Overhead Scheduling Scheme Based on Genetic Algorithm in Mobile Grids

Abstract

Mobile grid supports a new infrastructure to integrate and utilize the exploding population of mobile resources. The mobile grid is not as reliable as the traditional wired grid, mainly due to the users' movement. In this paper, we propose a job scheduling method based on Genetic Algorithm in mobile grid. The scheduling scheme takes into account the characteristics of mobile devices such as inferior reliability, limited capacity and capability. We adopt the replication scheme to enhance reliability of the job execution. Our proposed method minimizes the completion time of grid applications. Furthermore, our method efficiently utilizes mobile grid resources by avoiding the job executions in useless replica. The experimental results confirm that the proposed method outperforms the other conventional scheduling methods in terms of the completion time and wasted resources in job execution.