General Purpose computing on Graphics Processing Unit offers a remarkable speedup for data parallel workloads, leveraging GPUs computational power. However, differently from graphic computing, it requires highly reliable operation in several application domains. In this paper we present SIFI a reliability evaluation framework for soft-errors on AMD GPUs built on top of Multi2Sim, a micro-architectural level simulator. SIFI is capable of computing different reliability metrics by means of two different techniques: fault injection and ACE analysis. Experiments performed on a set of 14 GPGPU applications targeting the AMD Southern Islands GPU architecture show the capability of the tool and the potential of its use to support decisions about the best architectural parameters for a given application.

SIFI: AMD southern islands GPU microarchitectural level fault injector / Vallero, Alessandro; Gizopoulos, Dimitris; Di Carlo, Stefano. - STAMPA. - (2017), pp. 138-144. (Intervento presentato al convegno 23rd IEEE International Symposium on On-Line Testing and Robust System Design, IOLTS 2017 tenutosi a Hotel Makedonia Palace, Thessaloniki (Greece) nel 21 September 2017) [10.1109/IOLTS.2017.8046209].

SIFI: AMD southern islands GPU microarchitectural level fault injector

Vallero, Alessandro;Di Carlo, Stefano
2017

Abstract

General Purpose computing on Graphics Processing Unit offers a remarkable speedup for data parallel workloads, leveraging GPUs computational power. However, differently from graphic computing, it requires highly reliable operation in several application domains. In this paper we present SIFI a reliability evaluation framework for soft-errors on AMD GPUs built on top of Multi2Sim, a micro-architectural level simulator. SIFI is capable of computing different reliability metrics by means of two different techniques: fault injection and ACE analysis. Experiments performed on a set of 14 GPGPU applications targeting the AMD Southern Islands GPU architecture show the capability of the tool and the potential of its use to support decisions about the best architectural parameters for a given application.
2017
9781538603512
File in questo prodotto:
File Dimensione Formato  
IOLTS2017-SIFI-PostDraft.pdf

accesso aperto

Descrizione: Postprint draft
Tipologia: 2. Post-print / Author's Accepted Manuscript
Licenza: PUBBLICO - Tutti i diritti riservati
Dimensione 438.01 kB
Formato Adobe PDF
438.01 kB Adobe PDF Visualizza/Apri
Pubblicazioni consigliate

I documenti in IRIS sono protetti da copyright e tutti i diritti sono riservati, salvo diversa indicazione.

Utilizza questo identificativo per citare o creare un link a questo documento: https://hdl.handle.net/11583/2692801
 Attenzione

Attenzione! I dati visualizzati non sono stati sottoposti a validazione da parte dell'ateneo