On mapping and scheduling tasks with synchronization on clusters of machines

Abstract

In this work, a two-step approach is adopted for scheduling tasks with synchronous communication. To that end, an efficient algorithm, called GLB-Synch, is introduced for mapping clusters and ordering tasks on processors in one integrated step. The algorithm used the information obtained during the clustering step for selecting a cluster to be mapped on the least loaded processor. A performance study has been conducted on the GLB-Synch algorithm by simulation. We have shown by analysis and experimentation that the GLB-Sync algorithm retains the same low complexity cost of the first step for clustering. © Springer-Verlag 2004.