Logo image
Back
A self-adapting distributed memory package for fast signal transforms
Conference proceeding   Open access

A self-adapting distributed memory package for fast signal transforms

K Chen and J.R Johnson
18th International Parallel and Distributed Processing Symposium, 2004. Proceedings
2004
DOI: https://doi.org/10.1109/IPDPS.2004.1302961
url
http://citeseerx.ist.psu.edu/viewdoc/summary?doi=10.1.1.129.9082View
SubmittedOpen Access (License Unspecified),  Open

Abstract

Clustering algorithms Computer science Digital signal processing Distributed computing Fast Fourier transforms Matrix converters Packaging Prototypes Signal processing algorithms Tensile stress
Summary form only given. We present a self-adapting distributed memory package for computing the Walsh-Hadamard transform (WHT), a prototypical fast signal transform, similar to the fast Fourier transform. A family of distributed memory algorithms are derived from different factorizations of the WHT matrix. Different factorizations correspond to different data distributions and communication patterns. Thus, searching over the space of factorizations leads to the best data distribution and communication pattern for a given platform. The distributed memory WHT package provides a framework for converting factorizations of the WHT matrix into MPl programs and exploring their performance by searching the space of factorizations.

Metrics

2 Record Views

Details

Logo image