Gang scheduling in a two-cluster system implementing migrations and periodic feedback

Abstract

In order to maximize the efficiency of a complex distributed system such as a grid, a proper scheduling algorithm is necessary. The scheduling algorithm is responsible for allocating the available system resources to the existing jobs. In this paper, we consider scheduling gangs in a multi-cluster system in the presence of critical jobs which arrive in a sporadic manner. We examine an adaptive grid scheduler with periodic feedback which allocates the existing jobs to the available clusters. A migration scheme is implemented in order to alleviate the impact of critical sporadic jobs on the performance of gang scheduling. A simulation model is used to provide results on the performance of the system.