loading
Documents

Research.Publish.Connect.

Paper

Paper Unlock

Authors: Matthieu Martel ; Amine Najahi and Guillaume Revy

Affiliation: Univ. Perpignan Via Domitia, DALI, Univ. Montpellier II, LIRMM, UMR 5506, CNRS, LIRMM and UMR 5506, France

ISBN: 978-989-758-000-0

Keyword(s): Automated Code Synthesis, Matrix Multiplication, Fixed-point Arithmetic, Certified Numerical Accuracy.

Related Ontology Subjects/Areas/Topics: Embedded Communications Systems ; Pervasive Embedded Devices ; Software Architectures ; Telecommunications

Abstract: In digital signal processing, many primitives boil down to a matrix multiplication. In order to enable savings in time, energy consumption, and on-chip area, these primitives are often implemented in fixed-point arithmetic. Various conflicting goals undermine the process of writing fixed-point codes, such as numerical accuracy, runtime latency, and size of the codes. In this article, we introduce a new methodology to automate the synthesis of small and accurate codes for matrix multiplication in fixed-point arithmetic. Our approach relies on a heuristic to merge matrix rows or columns in order to reduce the synthesized code size, while guaranteeing a target accuracy. We suggest a merging strategy based on finding closest pairs of vectors, which makes it possible to address in a few seconds problems such as the synthesis of small and accurate codes for size-64 and more matrix multiplication. Finally, we illustrate its efficiency on a set of benchmarks, and we show that it allows to red uce the synthesized code size by more than 50% while maintaining good numerical properties. (More)

PDF ImageFull Text

Download
Sign In Guest: Register as new SciTePress user now for free.

Sign In SciTePress user: please login.

PDF ImageMy Papers

You are not signed in, therefore limits apply to your IP address 54.145.83.79

In the current month:
Recent papers: 100 available of 100 total
2+ years older papers: 200 available of 200 total

Paper citation in several formats:
Martel M., Najahi A. and Revy G. (2014). Code Size and Accuracy-aware Synthesis of Fixed-point Programs for Matrix Multiplication.In Proceedings of the 4th International Conference on Pervasive and Embedded Computing and Communication Systems - Volume 1: PECCS, ISBN 978-989-758-000-0, pages 204-214. DOI: 10.5220/0004884802040214

@conference{peccs14,
author={Matthieu Martel and Amine Najahi and Guillaume Revy},
title={Code Size and Accuracy-aware Synthesis of Fixed-point Programs for Matrix Multiplication},
booktitle={Proceedings of the 4th International Conference on Pervasive and Embedded Computing and Communication Systems - Volume 1: PECCS,},
year={2014},
pages={204-214},
publisher={SciTePress},
organization={INSTICC},
doi={10.5220/0004884802040214},
isbn={978-989-758-000-0},
}

TY - CONF

JO - Proceedings of the 4th International Conference on Pervasive and Embedded Computing and Communication Systems - Volume 1: PECCS,
TI - Code Size and Accuracy-aware Synthesis of Fixed-point Programs for Matrix Multiplication
SN - 978-989-758-000-0
AU - Martel M.
AU - Najahi A.
AU - Revy G.
PY - 2014
SP - 204
EP - 214
DO - 10.5220/0004884802040214

Login or register to post comments.

Comments on this Paper: Be the first to review this paper.