Gang scheduling in multi-core clusters implementing migrations

A proper scheduling algorithm is essential for the efficient utilization of the available resources of complex distributed systems. The scheduling algorithm is responsible for allocating the available system resources to the existing jobs. The emergence of multi-core processors poses new demands on schedulers. In this paper, we study the performance of gang scheduling algorithms for homogeneous and heterogeneous clusters which consist of multi-core processors. Furthermore, a migration schema is suggested which is suitable for scheduling gangs in multi-core clusters. A simulation model is used to provide results on the performance of the system.

► Proper scheduling algorithm promotes and enhances system performance. ► Performance benefits from the deployment of multi-core cpus in cluster computing. ► Gang scheduling is a special case of interacting tasks. ► Improved gang scheduling for multi-core clusters. ► Implementation of a migration schema for multi-core clusters.