Analysis of Synchronous Dynamic Load Balancing Algorithms

To share the work evenly among the processors is a crucial problem in order to fully exploit the parallel computers. The objective of any dynamic load balancing algorithm is to improve the overall performance of the machine by redistributing the work among the processors in order to minimize the program execution time. We proposed several load balancing schemes especially suited for data-parallel algorithms on SIMD computers. This paper presents the theoretical analysis of these load balancing strategies.

[1]  Vipin Kumar,et al.  Unstructured tree search on SIMD parallel computers: a summary of results , 1992, Proceedings Supercomputing '92.

[2]  Richard E. Korf,et al.  Depth-First Heuristic Search on a SIMD Machine , 1993, Artif. Intell..

[3]  Cyril Fonlupt,et al.  A Data-Parallel View of the Load Balancing - Experimental Results on MasPar MP-1 , 1994, HPCN.

[4]  Richard E. Korf,et al.  Parallel tree search on a SIMD machine , 1991, Proceedings of the Third IEEE Symposium on Parallel and Distributed Processing.

[5]  George Cybenko,et al.  Dynamic Load Balancing for Distributed Memory Multiprocessors , 1989, J. Parallel Distributed Comput..

[6]  Reinhard Lüling,et al.  Load balancing in large networks: a comparative study , 1991, Proceedings of the Third IEEE Symposium on Parallel and Distributed Processing.

[7]  W. Daniel Hillis,et al.  Data parallel algorithms , 1986, CACM.

[8]  Charles J. Daniels,et al.  A SIMD Approach to Parallel Heuristic Search , 1993, Artif. Intell..

[9]  Cyril Fonlupt.  Distribution dynamique de données sur machine SIMD , 1994 .