Auto-tuning for floating-point precision with Discrete Stochastic Arithmetic

•An algorithm and a software are proposed for floating-point precision auto-tuning.•Input programs are automatically modified taking into account accuracy requirements.•Discrete Stochastic Arithmetic is used to verify the results accuracy.•Performance results are presented on several benchmarks and...

Ausführliche Beschreibung

Gespeichert in:
Bibliographische Detailangaben
Veröffentlicht in:Journal of computational science Jg. 36; S. 101017
Hauptverfasser: Graillat, Stef, Jézéquel, Fabienne, Picot, Romain, Févotte, François, Lathuilière, Bruno
Format: Journal Article
Sprache:Englisch
Veröffentlicht: Elsevier B.V 01.09.2019
Elsevier
Schlagworte:
Auto-tuning
Computer Science
Discrete Stochastic Arithmetic
Floating-point arithmetic
Numerical validation
Round-off errors
Round-off errors
Numerical validation
Discrete Stochastic Arithmetic
Auto-tuning
Floating-point arithmetic
floating-point arithmetic
round-off errors
numerical validation
auto-tuning
ISSN:1877-7503, 1877-7511
Online-Zugang:Volltext
Tags: Tag hinzufügen
Keine Tags, Fügen Sie den ersten Tag hinzu!
Beschreibung
Zusammenfassung:•An algorithm and a software are proposed for floating-point precision auto-tuning.•Input programs are automatically modified taking into account accuracy requirements.•Discrete Stochastic Arithmetic is used to verify the results accuracy.•Performance results are presented on several benchmarks and on an industrial code. The type length chosen for floating-point numbers (e.g. 32 bits or 64 bits) may have an impact on the execution time, especially on SIMD (Single Instruction Multiple Data) units. Furthermore optimizing the types used in a numerical simulation causes a reduction of the data volume that is possibly transferred. In this paper we present PROMISE, a tool that makes it possible to optimize the numerical types in a program by taking into account the requested accuracy on the computed results. With PROMISE the numerical quality of results is verified using DSA (Discrete Stochastic Arithmetic) that enables one to estimate round-off errors. The search for a suitable type configuration is performed with a reasonable complexity thanks to the delta debugging algorithm. The PROMISE tool has been successfully tested on programs implementing several numerical algorithms including linear system solving and also on an industrial code that solves the neutron transport equations.
ISSN:1877-7503
1877-7511
DOI:10.1016/j.jocs.2019.07.004