A precedence based distributed job scheduling for computational grid

Scheduling on a computational grid is an NP Hard problem encouraging a number of scheduling strategies using various parameters and objectives. A computational grid provides the user a platform to execute the compute intensive jobs which otherwise cannot be executed at the user´s end. A grid system can be used to its full potential if the scheduling strategy provides the efficient mapping between the software parallelisms available in the application on the hardware parallelism offered by the grid. This work proposes a scheduling strategy that schedules the job on the suitable grid resources as per the job´s requirements while considering the communication requirements and preserving the precedence constraints within the job. Simulation study reveals the effectiveness of the model under various conditions.