A NEW-THRESHOLD BASED JOB SCHEDULING FOR GRID SYSTEM

Abstract

In heterogeneous distributed systems, utility grids have emerged as a new model of service. In this service, workflow scheduling is one of the challenging problems for satisfying user’s quality requirements. One of the main issues in work flow scheduling is to minimize the workflow execution cost in terms of time and makespan. In this study, we propose a new workflow scheduling algorithm based on a novel concept called New-Threshold Based Scheduling (NTBS) that attempts to minimize the cost of workflow execution time and provides service fairness. It works under two phases. The scheduler computes threshold in first phase and in second phase it schedules the grid jobs to reduce the execution time without affecting fairness. From the simulation results it is observed that NTBS gives better performance in terms of reduced makespan and consistent turnaround time as compared to FCFS, EDF and other scheduling algorithms.