The Community for Technology Leaders
Proceedings 21st International Conference on Distributed Computing Systems (2001)
Mesa, AZ
Apr. 16, 2001 to Apr. 19, 2001
ISBN: 0-7695-1077-9
pp: 0127
Ron Lavi , The Hebrew University of Jerusalem
Amnon Barak , The Hebrew University of Jerusalem
ABSTRACT
Abstract: Most implementations of a Computing Cluster (CC) use greedy-based heuristics to perform load balancing. In some cases, this is in contrast to theoretical results about the performance of on-line load balancing algorithms. In this paper we define the home model in order to better reflect the architecture of a CC. In this new theoretical model, we assume a realistic cluster structure in which every job has a "home" machine which it prefers to be executed on, e.g. due to I/O considerations or because it was created there. We develop several on-line algorithms for load balancing in this model. We first provide a theoretical worst-case analysis, showing that our algorithms achieve better competitive ratios and perform less reassignments than algorithms for the unrelated machines model, which is the best existing theoretical model to describe such clusters. We then present an empirical average-case performance analysis by means of simulations. We show that the performance of our algorithms is consistently better than that of several existing load balancing methods, e.g. the greedy and the opportunity cost methods, especially in a dynamic and changing CC environment.
INDEX TERMS
CITATION

R. Lavi and A. Barak, "The Home Model and Competitive Algorithms for Load Balancing in a Computing Cluster," Proceedings 21st International Conference on Distributed Computing Systems(ICDCS), Mesa, AZ, 2001, pp. 0127.
doi:10.1109/ICDSC.2001.918941
101 ms
(Ver 3.3 (11022016))