A Controlled Scheduling Algorithm Decreasing the Incidence of Starvation in Grid Environments

A fair scheduling algorithm accounting for the weight and execution time of tasks is critical in the Grid environment. The MTWCT (Minimize Total Weighted Completion Time) has been proved to minimize the total weighted completion time of a set of independent tasks in a processor, but it results in another problem that the response time of some tasks is far longer. To decrease the incidence of the starvation phenomena, an improved algorithm named CSA (Controlled Scheduling algorithm) based on MTWCT is proposed, which computes the ρ factors of tasks by the execution time and weight of step chains, and selects the unexecuted step chain in terms of ρ factor and the executed time of task. Experimental results exhibit that CSA compared with the MTWCT, decreases the completion time of short tasks and the average turnaround time by sacrificing a little total weighted completion time.