Dynamic task scheduling for irregular network topologies M-Tahar Kechadi * , Ilias K. Savvas Parallel Computational Research Group, Department of Computer Science, University College Dublin, Belfield, Dublin 4, Ireland Received 31 October 2004; received in revised form 15 March 2005; accepted 15 April 2005 Available online 23 June 2005 Abstract In this study, a heterogeneous computing environment is employed as a computational platform. In order to increase the efficiency of the system, a dynamic task-scheduling algo- rithm is proposed, which balances the load among the nodes of the system. The technique is dynamic, nonpreemptive, adaptive, and it uses a mixed centralised and decentralised poli- cies.Basedonthedivideandconquerprinciple,thealgorithmmodelsthesystemashypergrids and then balances the load among them. Recursively, the hypergrids of dimension k are dividedintogridsofdimensions k  1,untilthedimensionis1.Then,allthenodesofthesys- temarealmostequallyloaded.Theoptimumdimensionofthehypergridischoseninorderto achievethebestperformance.Thesimulationresultsdemonstratetheeffectivenessofthistech- nique. In addition, we determined the critical points representing lower bounds for which the algorithm should be effective and therefore should be triggered. Ó 2005 Elsevier B.V. All rights reserved. Keywords: Scheduling; Load-balancing; Heterogeneous distributed system; n-D grid 0167-8191/$ - see front matter Ó 2005 Elsevier B.V. All rights reserved. doi:10.1016/j.parco.2005.04.007 * Corresponding author. Tel.: +353 1 716 2478; fax: +353 1 269 7262. E-mail address: tahar.kechadi@ucd.ie (M-Tahar Kechadi). www.elsevier.com/locate/parco Parallel Computing 31 (2005) 757–776