Design and Evaluation of Agent Based Prioritized Dynamic Round Robin Scheduling Algorithm on Computational Grids

Grid computing enables sharing, selection and aggregation of computing resources for solving complex and large-scale scientific problems. Grid scheduling is playing a vital role for the efficient and effective execution of jobs on computational grids. Most of the scheduling algorithms do not consider user and system objectives at the same time. Therefore, in this paper we introduce a concept of fairness to scheduling and present a new agent based job scheduling algorithm called Agent based Prioritized Dynamic Round Robin (APDRR). APDRR is designed and developed by combining the best features of round robin and priority job scheduling algorithm using agent technology. APDRR is fair scheduling algorithm from a user point of view while also regarding the optimization criteria that are anticipated from the system perspective. This paper also presents the comparative performance analysis of our proposed APDRR along with other well known job scheduling algorithms, considering the performance metrics comprised of average waiting time, average turnaround time, average response time, total completion time, average bounded slowdown time and maximum job stretch time. Performance evaluation of job scheduling algorithms has been carried out on a computational grid using real workload traces. Experimental evaluation confirmed that the proposed APDRR scheduling algorithm posses a high degree of optimality in performance, efficiency and scalability. This paper also includes a statistical analysis of real workload traces to present the nature and behavior of jobs