Logo image
Back
SPIRAL: Code Generation for DSP Transforms
Journal article   Open access

SPIRAL: Code Generation for DSP Transforms

M Puschel, Y Voronenko, K Chen, R.W Johnson, N Rizzolo, J.M.F Moura, J.R Johnson, D Padua, M.M Veloso, B.W Singer, …
Proceedings of the IEEE, v 93(2), pp 232-275
Feb 2005
DOI: https://doi.org/10.1109/JPROC.2004.840306
url
http://citeseerx.ist.psu.edu/viewdoc/summary?doi=10.1.1.116.294View
Submitted,  Open

Abstract

Adaptation automatic performance tuning code optimization Digital signal processing discrete cosine transform (DCT) discrete Fourier transform (DFT) Discrete Fourier transforms Discrete wavelet transforms fast Fourier transform (FFT) filter Fourier transforms genetic and evolutionary algorithm high-performance computing Humans learning library generation linear signal transform Markov decision process Microarchitecture Portable computers Scientific computing search Signal processing algorithms Spirals wavelet
Fast changing, increasingly complex, and diverse computing platforms pose central problems in scientific computing: How to achieve, with reasonable effort, portable optimal performance? We present SPIRAL, which considers this problem for the performance-critical domain of linear digital signal processing (DSP) transforms. For a specified transform, SPIRAL automatically generates high-performance code that is tuned to the given platform. SPIRAL formulates the tuning as an optimization problem and exploits the domain-specific mathematical structure of transform algorithms to implement a feedback-driven optimizer. Similar to a human expert, for a specified transform, SPIRAL "intelligently" generates and explores algorithmic and implementation choices to find the best match to the computer's microarchitecture. The "intelligence" is provided by search and learning techniques that exploit the structure of the algorithm and implementation space to guide the exploration and optimization. SPIRAL generates high-performance code for a broad set of DSP transforms, including the discrete Fourier transform, other trigonometric transforms, filter transforms, and discrete wavelet transforms. Experimental results show that the code generated by SPIRAL competes with, and sometimes outperforms, the best available human tuned transform library code.

Metrics

17 Record Views
423 citations in Web of Science
634 citations in Scopus

Details

InCites Highlights

Data related to this publication, from InCites Benchmarking & Analytics tool:

Collaboration types
Industry collaboration
Domestic collaboration
Web of Science research areas
Engineering, Electrical & Electronic
Logo image