Abstract
This paper takes an experimental approach to the load balancing problem for parallel simulation applications. In particular, it focuses upon a conservative synchronization protocol, by making use of an optimized version of Chandy&Misra null message method, and propose a dynamic load balancing algorithm which assumes no compile time knowledge about the workload parameters. The proposed scheme is also implemented on an Intel Paragon A4 multicomputer, and the performance results for several simulation models are reported.
Part of this research is supported by Texas Advanced Technology Program grant TATP-003594031
Chapter PDF
Keywords
These keywords were added by machine and not by the authors. This process is experimental and the keywords may be updated as the learning algorithm improves.
References
Boukerche, A., Das, S. K.: Dynamic Load Balancing Strategies For Parallel Simulations. TR-98. University of North Texas.
Boukerche, A., and Tropper C., “A Static Partitioning and Mapping Algorithm for Conservative Parallel Simulations”, IEEE/ACM PADS’94, 1994, 164–172.
Misra, J., “Distributed Discrete-Event Simulation”, ACM Computing Surveys, Vol. 18, No. 1, 1986, 39–65.
Author information
Authors and Affiliations
Editor information
Rights and permissions
Copyright information
© 1998 Springer-Verlag Berlin Heidelberg
About this paper
Cite this paper
Boukerche, A., Das, S.K. (1998). Experimental studies in load balancing. In: Pritchard, D., Reeve, J. (eds) Euro-Par’98 Parallel Processing. Euro-Par 1998. Lecture Notes in Computer Science, vol 1470. Springer, Berlin, Heidelberg. https://doi.org/10.1007/BFb0057868
Download citation
DOI: https://doi.org/10.1007/BFb0057868
Published:
Publisher Name: Springer, Berlin, Heidelberg
Print ISBN: 978-3-540-64952-6
Online ISBN: 978-3-540-49920-6
eBook Packages: Springer Book Archive