Title :
Parallelization of the dynamic programming algorithm for the matrix chain product on a hypercube
Author :
Strate, Steve A. ; Wainwright, Roger L.
Author_Institution :
Dept. of Math. & Comput. Sci., Tulsa Univ., OK, USA
Abstract :
An examination is made of the dynamic programming technique implemented on a distributed-memory multiprocessor system. The matrix chain product algorithm is investigated as an example problem. A method for parallelizing the dynamic programming technique for solving the matrix chain product problem is presented and load-balancing considerations are examined. The sequential dynamic programming technique is a fine grain algorithm and considered by many researchers to be too fine grain to execute effectively on a hypercube. The parallel algorithm yielded modest speedups for fixed-size problems. Scaled problems promise even better speedups. Results show that respectable performance for the dynamic programming algorithmic technique can be achieved on a hypercube
Keywords :
dynamic programming; matrix algebra; parallel algorithms; parallel machines; algorithmic technique; distributed-memory multiprocessor system; dynamic programming algorithm; dynamic programming technique; fine grain algorithm; fixed-size problems; hypercube; load-balancing considerations; matrix chain product algorithm; modest speedups; parallel algorithm; parallelization; sequential dynamic programming; Computer science; Costs; Dynamic programming; Heuristic algorithms; Hypercubes; Load management; Matrix decomposition; Multiprocessing systems; Parallel algorithms; Parallel programming;
Conference_Titel :
Applied Computing, 1990., Proceedings of the 1990 Symposium on
Conference_Location :
Fayetteville, AR
Print_ISBN :
0-8186-2031-5
DOI :
10.1109/SOAC.1990.82144