Job Scheduling for Heterogeneous Computing Environments

• Published in: International journal of modeling and optimization Vol. 2; no. 6; pp. 697 - 700
• Main Authors: Devaki, P, Valarmathi, M L
• Format: Journal Article
• Language: English
• Published: 01.12.2012
• Subjects: Algorithms; Complexity; Computation; Computational efficiency; Dynamical systems; Dynamics; Queues; Scheduling
• ISSN: 2010-3697; 2010-3697
• DOI: 10.7763/IJMO.2012.V2.213

Scheduling the jobs of computationally intensive applications efficiently is one of the most essential and difficult issues when aiming at high performance in heterogeneous computing environments. A large number of scheduling heuristics have been presented in literature for homogeneous computing systems. The complexity of the problem increases when job scheduling is to be carried out in heterogeneous computing system. In this paper, we present a simple algorithm Heterogeneous Task Scheduling (HTS) for a bounded number of heterogeneous machines. The aim of our algorithm is to minimize the overall completion time of jobs (makespan) submitted. The methodology used is to maintain the dynamic queue (Ready Queue) in which the ready jobs are available. From that queue the job is selected for execution on a machine which is capable of completing that job quickly. The analysis and experiments have shown that this algorithm provides comparable results in some cases and even better results in most of the cases together with low complexity when compared with the existing algorithms Critical-Path On a Processor (CPOP) and Heterogeneous Critical Parent Trees (HCPT).