Task Load Balancing Strategy for Grid Computing

Abstract

Workload and resource management are two essential functions provided at the service level of the Grid software infrastructure. To improve the global throughput of these environments, effective and efficient load balancing algorithms are fundamentally important. Most strategies were developed in mind, assuming homogeneous set of resources linked with homogeneous and fast networks. However for computational Grids we must address main new challenges, like heterogeneity, scalability and adaptability. Our contributions in this perspective are two fold. First we propose a dynamic treebased model to represent Grid architecture in order to manage workload. This model was characterized by three main features: (i) it was hierarchical; (ii) it supports heterogeneity and scalability; and (iii) it was totally independent from any Grid physical architecture. Second, we develop a hierarchical load balancing strategy and associated algorithms based on neighbourhood propriety. The main benefit of this idea was to decrease the amount of messages exchanged between Grid resources. As consequence, the communication overhead induced by tasks transferring and flow information was reduced. In order to evaluate the practicability and performance of our strategy we have developed a Grid simulator in Java. The first results of our experimentations were very promising. We have realized a significant improvement in mean response time with a reduction of communication cost. It means that the proposed model can lead to a better load balancing between resources without high overhead.

3 Figures and Tables

Statistics

051020072008200920102011201220132014201520162017
Citations per Year

65 Citations

Semantic Scholar estimates that this publication has 65 citations based on the available data.

See our FAQ for additional information.

Cite this paper

@inproceedings{Yagoubi2007TaskLB, title={Task Load Balancing Strategy for Grid Computing}, author={Belabbas Yagoubi and Yacine Slimani}, year={2007} }