Method and system for integrated scheduling and replication in a grid computing system

摘要

A method for scheduling a plurality of computation jobs to a plurality of data processing units (DPUs) in a grid computing system 100. The method includes receiving a first computation job by a scheduling DPU from the plurality of computation jobs. Further, the method includes scheduling the first computation job at a first set of DPUs in the grid computing system based on first scheduling criteria. Furthermore, the method includes scheduling the first computation job at a first DPU from the first set of DPUs based on second scheduling criteria. The method also includes storing information about the first computation job and the first set of DPUs in a history table at the scheduling DPU. The method further includes storing information about the first DPU in a demand matrix of the first set of DPUs.