Resource Characteristic Based Optimization for Grid Scheduling

Scheduling is an active research area in the computational grid environment. The objective of grid scheduling is to deliver both the requirement of Quality of Service (QoS) by the grid users, as well as the requirement of high utilization of resources by the service providers. To obtain optimal scheduling in the generalized grid environment is an NP-complete problem. A large number of researchers have presented heuristic algorithms to find a near-global optimum for the static scheduling model of the grid. Relatively a smaller number of researchers have worked on the scheduling problem for the dynamic scheduling model. This paper proposes a new resource characteristic based optimization method, which may be combined with Earlier Gap, Earliest Deadline First (EG-EDF) policy to schedule jobs in a dynamic environment. The proposed algorithm generates near optimal solutions, which are better than those reported in the literature for a specific range of datasets. Extensive experimentation has proved the efficacy of our method.