 |
 |
|
|
| HAL : inria-00614597, version 1 |
| Voir la fiche détaillée |  BibTeX,EndNote,... |
 |
| Versions disponibles | v1 (12-08-2011) | v2 (07-12-2011) | v3 (09-04-2012) |
|
|
|
|
| Damaris: Leveraging Multicore Parallelism to Mask I/O Jitter |
|
|
| Matthieu Dorier 1Gabriel Antoniu 2 |
|
|
| Joint INRIA/UIUC Laboratory for Petascale Computing Collaboration(s) |
|
|
| (12/08/2011) |
|
|
|
| With exascale computing on the horizon, the performance variability of I/O systems presents a key challenge in sustaining high performance. In many HPC applications, I/O is performed concurrently by all processes; this produces I/O bursts, which causes resource contention and substantial variability of I/O performance, significantly impacting the overall application performance. In this paper, we utilize the IOR benchmark to explore the influence of user-configurable parameters on I/O variability. We then propose a new approach, called Damaris, leveraging dedicated I/O cores on each multicore SMP node to efficiently perform asynchronous data processing and I/O. We evaluate our approach on two different platforms with the CM1 atmospheric model, one of the BlueWaters HPC applications. By gathering data into large files while avoiding synchronization between cores, our solution increases the I/O throughput by a factor of 6, hides all I/O-related costs, and enables a 600\% compression ratio without any additional overhead. |
|
|
|
|
|
|
|
|
|
|
|
| 1 : | École normale supérieure de Cachan, antenne de Bretagne (ENS Cachan Bretagne) |
|
|
|
École normale supérieure de Cachan - ENS Cachan |
|
|
| 2 : | KerData (INRIA - IRISA) |
|
|
|
INRIA – CNRS : UMR6074 – École normale supérieure de Cachan - ENS Cachan – Institut National des Sciences Appliquées (INSA) - Rennes – Université de Rennes 1 |
|
|
| 3 : | GRAND-LARGE (INRIA Saclay - Ile de France) |
|
|
|
INRIA – CNRS : UMR8623 – Université Paris XI - Paris Sud |
|
|
| 4 : | Joint Laboratory for Petascale Computing [Illinois] (JLPC) |
|
|
|
University of Illinois at Urbana-Champaign – INRIA |
|
|
| 5 : | Laboratoire de Recherche en Informatique (LRI) |
|
|
|
CNRS : UMR8623 – Université Paris XI - Paris Sud |
|
|
| 6 : | Department of Computer Science [UIUC] (UIUC) |
|
|
|
University of Illinois at Urbana-Champaign |
|
|
| 7 : | University of Michigan |
|
|
|
University of Michigan |
|
|
|
|
|
|
|
|
|
| Domaine | : | Informatique/Calcul parallèle, distribué et partagé |
|
|
| Exascale Computing – Multicore Architectures – I/O – Variability – Dedicated Cores |
|
|
|
| Liste des fichiers attachés à ce document : | ||||||||||
|
||||||||||
|
|
|
| inria-00614597, version 1 | |
| http://hal.inria.fr/inria-00614597 | |
| oai:hal.inria.fr:inria-00614597 | |
| Contributeur : Matthieu Dorier | |
| Soumis le : Vendredi 12 Août 2011, 15:43:58 | |
| Dernière modification le : Mercredi 17 Août 2011, 12:59:42 | |
