Share Email Print

Proceedings Paper

Linda implementation on the ICAP level of the image-understanding architecture
Author(s): Sunit K. Bhalla; Charles C. Weems
Format Member Price Non-Member Price
PDF $14.40 $18.00
cover GOOD NEWS! Your organization subscribes to the SPIE Digital Library. You may be able to download this paper for free. Check Access

Paper Abstract

Using the Linda system, a parallel matrix multiplication program was written. This program uses N2 + 1 ICAP processors to compute the product of two N x N matrices. Due to the computational simplicity of matrix multiplication, a single ICAP processor running a sequential matrix multiplication program was faster than the Linda program. The computational complexity for which the Linda system becomes faster than a single ICAP processor was found. This was done by assuming that N2 computations still needed to be done, each requiring the row of one matrix and the column of another. Once a processor acquires these two arrays, an operation was simulated by running a delay loop. By increasing the size of the delay loop, more complex operations could be "simulated."

Paper Details

Date Published: 1 July 1990
PDF: 11 pages
Proc. SPIE 1246, Parallel Architectures for Image Processing, (1 July 1990); doi: 10.1117/12.19583
Show Author Affiliations
Sunit K. Bhalla, Hewlett-Packard Co. (United States)
Charles C. Weems, Univ. of Massachusetts (United States)

Published in SPIE Proceedings Vol. 1246:
Parallel Architectures for Image Processing
Joydeep Ghosh; Colin G. Harrison, Editor(s)

© SPIE. Terms of Use
Back to Top