Performance Comparison of Some Hybrid Deadline Based Scheduling Algorithms for Computational Grid

Grid computing is a form of distributed computing that involves collection of independent computers coordinating and sharing computing, application, data storage or network resources with high speed across dynamic and geographically distributed environment. Grid infrastructure plays a vital role in terms of computation in the performance call center. Moreover, grid scheduling is a vital component of a Computational Grid infrastructure. Typical scheduling challenges tend to be NP-hard problems where there is no optimal solution. In this paper, we proposed and evaluate few hybrid scheduling algorithms (Least Slack Time Round Robin Based Scheduling Algorithm (LSTRR), Shortest Processing Time First Round Robin Based Scheduling Algorithm (SPTFRR), Earliest Deadline First Round Robin Based Scheduling Algorithm (EDFRR) and Firs Come First Served Scheduling Algorithm (FCFS) ) based on deadline, slack time and baseline approaches for a real grid environment using real workload traces, taken from leading computational centers. An extensive performance comparison is presented using real workload traces to evaluate the efficiency of scheduling algorithms. Moreover, experimental results, based on performance metrics, demonstrate that the performances of our grid scheduling algorithms give good results. Our proposed schedule algorithms also support true scalability, that is, they maintain an efficient approach when increasing the number of processors on a real grid environment.