Paper
20 April 1988 Efficient Bit-Level, Word-Level, And Block-Level Systolic Arrays For Matrix-Matrix Multiplication
A J De Groot, S R Parker, E M Johansson
Author Affiliations +
Proceedings Volume 0880, High Speed Computing; (1988) https://doi.org/10.1117/12.944043
Event: 1988 Los Angeles Symposium: O-E/LASE '88, 1988, Los Angeles, CA, United States
Abstract
This paper investigates the mapping of matrix-matrix multiplication onto bit level, word level and block level systolic arrays. Highly efficient and regular bit level, word level and block level systolic arrays are described. Efficiencies of many block level and word level systolic arrays reported in this paper approach 100%, three times the efficiencies of systolic arrays reported previously. Bit level systolic arrays reported in this paper require less computation time than than do bit level systolic arrays reported previously and, for special matrices, require less cells. Execution times of block level systolic algorithms on a sixty-four-element multiprocessor agree with theory.
© (1988) COPYRIGHT Society of Photo-Optical Instrumentation Engineers (SPIE). Downloading of the abstract is permitted for personal use only.
A J De Groot, S R Parker, and E M Johansson "Efficient Bit-Level, Word-Level, And Block-Level Systolic Arrays For Matrix-Matrix Multiplication", Proc. SPIE 0880, High Speed Computing, (20 April 1988); https://doi.org/10.1117/12.944043
Lens.org Logo
CITATIONS
Cited by 5 patents.
Advertisement
Advertisement
RIGHTS & PERMISSIONS
Get copyright permission  Get copyright permission on Copyright Marketplace
KEYWORDS
Matrices

Algorithm development

Algorithms

Clocks

Array processing

Matrix multiplication

Tin

RELATED CONTENT

Implementation Of Cellular Arrays
Proceedings of SPIE (July 30 1982)
Scheduling Linearly Indexed Assignment Codes
Proceedings of SPIE (May 17 1989)
Scattering Arrays For Matrix Computations
Proceedings of SPIE (July 30 1982)
On Programming Languages for VLSI Array Processors
Proceedings of SPIE (July 28 1986)

Back to Top