Abstract: The proper functioning of a grid depends mainly on the efficient management and use of grid resources to carry out the various jobs that users send for execution. In grid computing system management of resources and scheduling of jobs are the most crucial problem, a lot of effort had been made to efficiently schedule and manage resource in computational grid systems, scheduling policies such as, First Come First Serve which is a simple policy used to improve efficiency of the scheduler, even though it have been widely used, it therefore end up with invoking low system utilization of both global and local scheduler as a result of the gaps that exists in between the jobs submitted in the waiting queue. To improve system utilization and efficiency of both main and local schedulers back filling approach is used, this approach allows tasks or jobs in the waiting queue to fill in those gaps that exist. This paper proposes an efficient scheduling algorithm based on multi-gap elimination approach. The proposed algorithm uses intelligent agents in the scheduler to perform scheduling in a collaborative and coordinated way. The algorithm uses gap filling strategy to optimize priority rule algorithms in grid scheduling system by considering the available gaps in both global and local schedulers.

Keywords: Gap filling, Scheduler, Algorithm, Task.