Reconfigurable arithmetic for HPC

Florent De Dinechin 1, 2 Bogdan Pasca 3
2 ARIC - Arithmetic and Computing
Inria Grenoble - Rhône-Alpes, LIP - Laboratoire de l'Informatique du Parallélisme
Abstract : An often overlooked way to increase the efficiency of HPC on FPGA is to tailor, as tightly as possible, the arithmetic to the application. An ideally efficient implementation would, for each of its operations, toggle and transmit just the number of bits required by the application at this point. Conventional microprocessors, with their word-level granularity and fixed memory hierarchy, keep us away from this ideal. FPGAs, with their bit-level granularity, have the potential to get much closer. Therefore, reconfigurable computing should systematically investigate, in an application-specific way, non-standard precisions, but also non-standard number systems and non-standard arithmetic operations. The purpose of this chapter is to review these opportunities.
Type de document :
Chapitre d'ouvrage
Wim Vanderbauwhede and Khaled Benkrid. High-Performance Computing using FPGAs, Springer, 2013
Liste complète des métadonnées

Littérature citée [74 références]  Voir  Masquer  Télécharger

https://hal.inria.fr/ensl-00758377
Contributeur : Florent De Dinechin <>
Soumis le : mercredi 28 novembre 2012 - 16:05:07
Dernière modification le : vendredi 20 avril 2018 - 15:44:26
Document(s) archivé(s) le : samedi 17 décembre 2016 - 16:35:01

Fichier

2012-Arith4HPC-chapter.pdf
Fichiers éditeurs autorisés sur une archive ouverte

Identifiants

  • HAL Id : ensl-00758377, version 1

Collections

Citation

Florent De Dinechin, Bogdan Pasca. Reconfigurable arithmetic for HPC. Wim Vanderbauwhede and Khaled Benkrid. High-Performance Computing using FPGAs, Springer, 2013. 〈ensl-00758377〉

Partager

Métriques

Consultations de la notice

275

Téléchargements de fichiers

279