SMCV: a Methodology for Detecting Transient Faults in Multicore Clusters

The challenge of improving the performance of current processors is achieved by increasing the integration scale. This carries a growing vulnerability to transient faults, which increase their impact on multicore clusters running large scientific parallel applications. The requirement for enhancing...

Full description

Saved in:
Bibliographic Details
Published inCLEI electronic journal Vol. 15; no. 3; p. 5
Main Authors Montezanti, Diego, Frati, Fernando Emmanuel, Rexachs, Dolores, Luque, Emilio, Naiouf, Marcelo, De Giusti, Armando
Format Journal Article
LanguageEnglish
Portuguese
Published Centro Latinoamericano de Estudios en Informática 01.12.2012
Subjects
COMPUTER SCIENCE, ARTIFICIAL INTELLIGENCE
COMPUTER SCIENCE, CYBERNETICS
COMPUTER SCIENCE, HARDWARE & ARCHITECTURE
COMPUTER SCIENCE, INFORMATION SYSTEMS
COMPUTER SCIENCE, INTERDISCIPLINARY APPLICATIONS
COMPUTER SCIENCE, SOFTWARE ENGINEERING
COMPUTER SCIENCE, THEORY & METHODS
ENGINEERING, ELECTRICAL & ELECTRONIC
cluster de multicores
reliability
transient fault
multicore cluster
aplicación científica paralela
corrupción silenciosa de datos
silent data corruption
message content validation
fiabilidad
detección de soft errors
parallel scientific application
fallo transitorio
validación de contenidos de  mensajes
soft error detection
Online AccessGet full text

Cover

More Information
Summary:The challenge of improving the performance of current processors is achieved by increasing the integration scale. This carries a growing vulnerability to transient faults, which increase their impact on multicore clusters running large scientific parallel applications. The requirement for enhancing the reliability of these systems, coupled with the high cost of rerunning the application from the beginning, create the motivation for having specific software strategies for the target systems. This paper introduces SMCV, which is a fully distributed technique that provides fault detection for message-passing parallel applications, by validating the contents of the messages to be sent, preventing the transmission of errors to other processes and leveraging the intrinsic hardware redundancy of the multicore. SMCV achieves a wide robustness against transient faults with a reduced overhead, and accomplishes a trade-off between moderate detection latency and low additional workload.
ISSN:0717-5000
0717-5000
DOI:10.19153/cleiej.15.3.5