Recovery-Restart Techniques for Resilient Krylov Methods

Abstract : he advent of extreme scale machines will require the use of parallel resources at an unprecedented scale, possibly leading to a high rate of hardware faults. In this presentation, we investigate recovery followed by restarting strategies for the resilience of Krylov subspace linear solvers as well as of eigensolvers.
Type de document :
Communication dans un congrès
SIAM Conference on Parallel Processing for Scientific Computing (SIAM PP 2014), Feb 2014, Portland, United States. 2014
Liste complète des métadonnées

https://hal.inria.fr/hal-01062232
Contributeur : Mawussi Zounon <>
Soumis le : mardi 9 septembre 2014 - 14:46:59
Dernière modification le : lundi 22 octobre 2018 - 16:56:01

Identifiants

  • HAL Id : hal-01062232, version 1

Citation

Emmanuel Agullo, Luc Giraud, Salas Pablo, Mawussi Zounon. Recovery-Restart Techniques for Resilient Krylov Methods. SIAM Conference on Parallel Processing for Scientific Computing (SIAM PP 2014), Feb 2014, Portland, United States. 2014. 〈hal-01062232〉

Partager

Métriques

Consultations de la notice

234