A Trust Based Job Scheduling Algorithm in Grid Network

Abstract

The Computational grid provides promising platform for efficient execution of resource intensive parameter sweep applications
or jobs. But the parameter sweep jobs may or may not execute within the estimated execution time or estimated resource requirements.
Sometimes the jobs can hold the resources longer time than the estimated time. It increases the turnaround time of other jobs that always
execute within the estimated execution time. This imposes a challenge of scheduling grid jobs in an efficient way so that neither of the jobs
starves. Although numerous works have been done in grid scheduling, no grid scheduling algorithm has considered all the factors like trust,
resource need, priority and time consumption. To address this problem, a trust based job scheduling algorithm has been proposed and
developed. The proposed approach considers all the factors aforementioned to calculate Scheduling factor of a job and the jobs can be
scheduled on the basis of Scheduling factor. An empirical study has also been carried out in order to prove the significance of proposed
approach and results shows that the turnaround time of trustworthy jobs can be reduced significantly by more than 50%.

Keywords: Grid Computing; Grid Scheduling; Trust; Job scheduling; Grid system;