Logo image
A prototypical self-optimizing package for parallel implementation of fast signal transforms
Conference proceeding   Open access

A prototypical self-optimizing package for parallel implementation of fast signal transforms

Kang Chen, J. R Johnson and Kenneth Chen
Proceedings 16th International Parallel and Distributed Processing Symposium, pp 6 pp-548
2002
url
http://www.cs.drexel.edu/~jjohnson/2007-08/fall/cs680/papers/ipdps02.pdfView

Abstract

Computer science Concurrent computing Digital signal processing Fast Fourier transforms Mathematics Packaging Partitioning algorithms Prototypes Signal processing algorithms Tensile stress
This paper presents a sell-adapting parallel package for computing the Walsh-Hadamard transform (WHT), a prototypical fast signal transform, similar to the fast Fourier transform. Using a search over a space of mathematical formulas representing different algorithms to compute the WHT the package finds the best parallel implementation on a given shared-memory multiprocessor. The search automatically finds the best combination of sequential and parallel code leading to the most effective granularity, load balance and cache utilization. Experimental results are presented showing the optimizations required to obtain nearly linear speedup on a sample symmetric multiprocessor.

Metrics

11 Record Views
3 citations in Scopus

Details

Logo image