Predicting robustness against transient faults of MPI based programs

The evaluation of a program's behaviour in the presence of transient faults is often a very time consuming work. In order to achieve significant data, thousands of executions are required and each execution will have the significant overhead of the fault injection environment. A previously publ...

Descripción completa

Detalles Bibliográficos
Autores: Dias Lima Gramacho, João Artur, Wong, Álvaro|||0000-0002-8394-9478, Rexachs, Dolores|||0000-0001-5500-850X, Luque, Emilio|||0000-0002-2884-3232
Tipo de recurso: artículo
Fecha de publicación:2016
País:España
Institución:Universitat Autònoma de Barcelona
Repositorio:Dipòsit Digital de Documents de la UAB
Idioma:inglés
OAI Identifier:oai:ddd.uab.cat:160449
Acceso en línea:https://ddd.uab.cat/record/160449
https://dx.doi.org/urn:doi:10.1504/IJCSE.2016.076218
Access Level:acceso abierto
Palabra clave:Transient faults
Robustness prediction
Soft errors
Reliability
Parallel application signature
Performance prediction
PAS2P
MPI
Message passing interface
Program execution
Parallel programs
Descripción
Sumario:The evaluation of a program's behaviour in the presence of transient faults is often a very time consuming work. In order to achieve significant data, thousands of executions are required and each execution will have the significant overhead of the fault injection environment. A previously published methodology reduced significantly the time needed to evaluate the robustness of a program execution by exhaustively analysing its execution trace instead of using fault injection. In this paper we present a further improvement in the evaluation time of parallel programs robustness against transient faults by combining this methodology with PAS2P - a method that strives to describe an application based on its message-passing activity. This combination allowed us to predict the robustness of larger parallel programs, reducing in some cases by more than 20 times the time needed to calculate the robustness while obtaining a robustness prediction error of less than 4%.