An efficient grid scheduling algorithm with fault tolerance and user satisfaction

Abstract

Problem Statement. The advances in human civilization lead to more complications in problem solving. Grid computing serves as an efficient technology in solving those complicated problems. In computational grids, the grid scheduler schedules the task and finds the appropriate resource for each task. The scheduler must consider several factors such as user demand, communication time, failure handling mechanisms, and reduced makespan. Most of the existing algorithms do not consider user satisfaction. Thus a scheduling algorithm that handles failure of resources and achieves user satisfaction gains more importance. Approach. A new bicriteria scheduling algorithm (BSA) that considers user satisfaction along with fault tolerance has been introduced. The main contribution of this paper includes achieving user satisfaction along with fault tolerance and minimizing the makespan of jobs. Results. The performance of this proposed algorithm is evaluated using GridSim based on makespan and number of jobs completed successfully within user deadline. Conclusions/ Recommendations. The proposed BSA algorithm achieves reduced makespan and better hit rate with higher user satisfaction and fault tolerance. © 2013 P. Keerthika and N. Kasthuri.