Title :
A tensor product formulation of Strassen´s matrix multiplication algorithm with memory reduction
Author :
Kumar, B. ; Huang, C.-H. ; Johnson, R.W. ; Sadayappan, P.
Author_Institution :
Dept. of Comput. & Inf. Sci., Ohio State Univ., Columbus, OH, USA
Abstract :
A programming methodology based on tensor products has been used for designing and implementing block recursive algorithms for parallel and vector multiprocessors. A previous tensor product formulation of Strassen´s matrix multiplication algorithm requires working arrays of size O(7n) for multiplying 2n×2n matrices. The authors present a modified tensor product formulation of Strassen´s algorithm in which the size of working arrays can be reduced to O(4n). The modified formulation exhibits sufficient parallel and vector operations for efficient implementation. Performance results on the Cray Y-MP are presented
Keywords :
computational complexity; matrix algebra; parallel algorithms; Cray Y-MP; Strassen algorithm; block recursive algorithms; matrix multiplication algorithm; memory reduction; programming methodology; spare complexity; tensor product formulation; vector multiprocessors; working arrays; Algorithm design and analysis; Cloud computing; Computerized monitoring; Information science; Linear algebra; NIST; Parallel machines; Parallel programming; Tensile stress; Vectors;
Conference_Titel :
Parallel Processing Symposium, 1993., Proceedings of Seventh International
Conference_Location :
Newport, CA
Print_ISBN :
0-8186-3442-1
DOI :
10.1109/IPPS.1993.262814