Scheduling hybrid divisible and indivisible loads on clusters

Abstract

Divisible load applications occur in many fields of science and engineering. Such applications can be easily parallelized in a master-worker fashion, but pose several scheduling challenges. In this paper we investigate the scheduling problem for processing a set of tasks comprising both divisible and indivisible tasks on cluster systems. Indivisible loads are characterized by the property that they need to be processed on their entirety on a single processor while divisible loads can be distributed across several processing nodes by fully exploiting the underlying data parallelism. We propose a novel scheduling algorithm referred to as Hybrid Load Scheduling (HLS) algorithm for handling a set of tasks comprising both divisible and indivisible tasks on clusters. HLS fully exploits the parallelism in divisible loads to achieve high resource utilization. It also utilizes a pipelining technique to hide communication time and thus further optimize the overall performance. Simulations are presented to evaluate and compare the performance of the proposed strategy.