Controlling the Memory Subscription of Distributed Applications with a Task-Based Runtime System

Abstract : This talk discuss the need for cooperation between a task-based distributed application code and a run-time system engine to control the memory subscription levels throughout the execution. We show that the task paradigm allows to control the memory footprint of the application by throttling the task submission flow rate, striking a compromise between the performance benefits of anticipative task submission and the resulting memory consumption. We illustrate the benefits of our contribution on a compressed dense linear algebra distributed application.
Type de document :
Communication dans un congrès
SIAM Conference on Parallel Processing for Scientific Computing (SIAM PP 2016), Apr 2016, Paris, France. pp.318 - 327, 2016, 〈http://www.siam.org/meetings/pp16/〉
Liste complète des métadonnées

https://hal.inria.fr/hal-01380126
Contributeur : Marc Sergent <>
Soumis le : mercredi 12 octobre 2016 - 15:04:07
Dernière modification le : jeudi 11 janvier 2018 - 06:27:21
Document(s) archivé(s) le : samedi 4 février 2017 - 20:02:59

Fichier

siampp.pdf
Fichiers produits par l'(les) auteur(s)

Identifiants

  • HAL Id : hal-01380126, version 1

Collections

Citation

Marc Sergent, David Goudin, Samuel Thibault, Olivier Aumage. Controlling the Memory Subscription of Distributed Applications with a Task-Based Runtime System. SIAM Conference on Parallel Processing for Scientific Computing (SIAM PP 2016), Apr 2016, Paris, France. pp.318 - 327, 2016, 〈http://www.siam.org/meetings/pp16/〉. 〈hal-01380126〉

Partager

Métriques

Consultations de la notice

351

Téléchargements de fichiers

44