Salsa: An Automatic Tool to Improve the Numerical Accuracy of Programs

14 pages•Published: April 16, 2018

Abstract

This article describes Salsa, an automatic tool to improve the accuracy of the floating- point computations done in numerical codes. Based on static analysis methods by abstract interpretation, our tool takes as input an original program, applies to it a set of transformations and then generates an optimized program which is more accurate than the initial one. The original and the transformed programs are written in the same imperative language. This article is a concise description of former work on the techniques implemented in Salsa, extended with a presentation of the main software architecture, the inputs and outputs of the tool as well as experimental results obtained by applying our tool on a set of sample programs coming from embedded systems and numerical analysis.

Keyphrases: automatic transformation of programs, floating point arithmetic, numerical accuracy, static analysis

In: Natarajan Shankar and Bruno Dutertre (editors). Automated Formal Methods, vol 5, pages 63-76.

Links:	https://easychair.org/publications/paper/x58n
	https://doi.org/10.29007/j2fd

BibTeX entry

@inproceedings{AFM17:Salsa_Automatic_Tool_Improve,
  author    = {Nasrine Damouche and Matthieu Martel},
  title     = {Salsa: An Automatic Tool to Improve the Numerical Accuracy of Programs},
  booktitle = {Automated Formal Methods},
  editor    = {Natarajan Shankar and Bruno Dutertre},
  series    = {Kalpa Publications in Computing},
  volume    = {5},
  publisher = {EasyChair},
  bibsource = {EasyChair, https://easychair.org},
  issn      = {2515-1762},
  url       = {/publications/paper/x58n},
  doi       = {10.29007/j2fd},
  pages     = {63-76},
  year      = {2018}}

Download PDF Open PDF in browser