Performances VS Reliability: how to exploit Approximate Computing for Safety-Critical applications

Gennaro Rodrigues; Fernanda Lima Kastensmidt; Vincent Pouget; Alberto Bosio

doi:10.1109/IOLTS.2018.8474122

Retour

Performances VS Reliability: how to exploit Approximate Computing for Safety-Critical applications

Acte de colloque

Open Access

Performances VS Reliability: how to exploit Approximate Computing for Safety-Critical applications

Gennaro Rodrigues, Fernanda Lima Kastensmidt, Vincent Pouget et Alberto Bosio

IOLTS 2018 - 24th International Symposium on On-Line Testing and Robust System Design, pp.291-294

IOLTS 2018 - 24th International Symposium on On-Line Testing and Robust System Design (Platja d'Aro, Spain, 02/07/2018–04/07/2018)

2018

DOI: https://doi.org/10.1109/IOLTS.2018.8474122

Résumé

Fault tolerance

Approximate computing

Laser

Fault injection

Approximate Computing (AxC) paradigm aims at designing energy-efficient systems, saving computational resources , and presenting better execution times. AxC aims to selectively violate the specifications, trading accuracy off for efficiency. It has been demonstrated in the literature the effectiveness of imprecise computation for both software and hardware components implementing inexact algorithms, showing an inherent resiliency to errors. On the other hand, the hidden cost of AxC is the reduction on the inherent resiliency to errors of an application. This paper aims at analyzing the impact of AxC on the reliability.

Fichiers et liens (1)

url

Find in HALAfficher

Indicateurs

1 Consultations de la notice

Détails

Titre: Performances VS Reliability: how to exploit Approximate Computing for Safety-Critical applications
Créateurs - sans rôle: Gennaro Rodrigues - Universidade Federal do Rio Grande do Sul
Fernanda Lima Kastensmidt - Universidade Federal do Rio Grande do Sul
Vincent Pouget - Université de Montpellier, Institut d'Electronique et des Systèmes - IES
Alberto Bosio - Test and dEpendability of microelectronic integrated SysTems
Détails de publication: IOLTS 2018 - 24th International Symposium on On-Line Testing and Robust System Design, pp.291-294
Colloque: IOLTS 2018 - 24th International Symposium on On-Line Testing and Robust System Design (Platja d'Aro, Spain, 02/07/2018–04/07/2018)
Identifiants: 9945496609311
Unité académique: Institut d'Electronique et des Systèmes - IES; Laboratoire d'Informatique de Robotique et de Microélectronique de Montpellier - LIRMM
Langue: English
Type de ressource: Conference proceeding
Champs locaux: hal-02095642