Abstract
This paper describes an adaptive algorithm called ALBA (Adaptive Load Balancing Algorithm) for load balancing of irregular applications on parallel and distributed architectures. ALBA uses an adaptive centralized load information collection policy to maintain a global current load state in the machine. The transfer and location decisions are taken locally on each node of the architecture. A thread-based approach is used to implement ALBA on a farm of DEC/ALPHA processors. An application to a multithreaded IDA* program applied to the 15-puzzle problem is presented. An average speed-up of 14.8 has been obtained on the longest instances of the problem on a farm of 16 processors.
Preview
Unable to display preview. Download preview PDF.
References
S. Dowaji. Contribution à l'étude des problèmes d'équilibrage de charge dans les environnements distribués. PhD thesis, Université de Versailles, 1995.
D. Eager, E. Lazowska, and J. Zahorjan. Adaptive Load Sharing In Homogeneous Distributed Systems. IEEE Transactions on Software Engeneering, SE-12(No.5):662–675, 1986.
A. Geist, A. Beguelin, J. Dongarra, and al., editors. PVM: Parallel Virtual Machine, A User's guide and tutorial for Networked Parallel Computing. MIT Press, 1994.
Z. Hafidi. Parcours parallèle IDA* sous PVM. Mémoire de DEA. Université de Lille I (LIFL), 1994.
Z. Hafidi, E.G. Talbi, and G. Goncalves. Load balancing and parallel tree search: The MPIDA* algorithm. Parco'95 proc. Gent Belgium, Sept 1995.
R. E. Korf. Depth-first iterative-deepening: An optimal admissible tree search. Artificial Intelligence, Vol. 32(No. 27):97–109, Feb 1985.
N. Melab. Synthèse des méthodes de distribution statique et dynamique de la charge sur architectures MIMD. Etude de cas: langages fonctionnels. Publication interne. Univ. de Lille I, LIFL, (AS-94-160), Oct 1994.
N. Melab, N. Devesa, M.P. Lecouffe, and B. Toursel. An Adaptive Load Information Collection Policy. Int. Conf. on Parallel and Distributed Processing Techniques and Applications (PDPTA'96), Sunnyvale, California, USA, 9–11 Aug 1996.
R. Namyst and J.F. Méhaut. PM2: Parallel Multithreaded Machine. A computing environment for distributed architectures. Parco'95 proc. Gent Belgium, Sept 1995.
N.J. Nilsson. Principles of artificial intelligence. Palo Alto, CA:Tioga, 1980.
E. G. Talbi. Allocation dynamique de processus dans les systèmes distribués et parallèles: Etat de l'art. Publication interne. Univ. de Lille I, LIFL, (AS-95-162), Jan 1995.
J. Xu and K. Hwang. Heuristic Methods for Dynamic Load Balancing in A Message-Passing Supercomputer. IEEE, pages 888–897, Apr 1991.
Author information
Authors and Affiliations
Editor information
Rights and permissions
Copyright information
© 1996 Springer-Verlag Berlin Heidelberg
About this paper
Cite this paper
Melab, N., Devesa, N., Lecouffe, M.P., Toursel, B. (1996). Adaptive load balancing of irregular applications a case study: IDA* applied to the 15-puzzle problem. In: Ferreira, A., Rolim, J., Saad, Y., Yang, T. (eds) Parallel Algorithms for Irregularly Structured Problems. IRREGULAR 1996. Lecture Notes in Computer Science, vol 1117. Springer, Berlin, Heidelberg. https://doi.org/10.1007/BFb0030124
Download citation
DOI: https://doi.org/10.1007/BFb0030124
Published:
Publisher Name: Springer, Berlin, Heidelberg
Print ISBN: 978-3-540-61549-1
Online ISBN: 978-3-540-68808-2
eBook Packages: Springer Book Archive