Deadline scheduling with priority for client-server systems on the grid

Abstract

We present algorithms for the scheduling sequential tasks on a network enabled server (NES) environment. This article is an extension of the paper: "A Study of Deadline Scheduling for Client-Server Systems on the Computational Grid" by Takefusa et al. We mainly discuss a deadline scheduling with priority strategy that is more appropriate for multiclient, multiserver case. Importance is first given to the task's priority and then the task is allocated to the server that can meet the task's deadline. This may cause that some already allocated tasks on the server miss their deadline. We also augment the benefits of scheduling algorithms with load measurements (which is done with the use of a forecasting tool called FAST) and fallback mechanisms. The experimental results show that the deadline scheduling with priority along with fallback mechanism can increase the overall number of tasks executed by the NES.