GBFS: Efficient Data-Sharing on Hybrid Platforms. Towards adding WAN-Wide elasticity to DFSes.

Adrien Lèbre 1, 2 Gustavo Bervian Brand 1, 2
1 ASCOLA - Aspect and composition languages
LINA - Laboratoire d'Informatique de Nantes Atlantique, Département informatique - EMN, Inria Rennes – Bretagne Atlantique
Abstract : Applications dealing with huge amounts of data suffer significant performance impacts when they are deployed on top of an hybrid platform (i.e the extension of a local infras- tructure with external cloud resources). More precisely, through a set of preliminary experiments we show that mechanisms which enable on demand extensions of current Distributed File Systems (DFSes) are required. These mechanisms should be able to leverage external storage resources while taking into account the performance constraints imposed by the physical network topology used to interconnect the different sites. Our answer to such a challenge is the Group Based File System proposal (GBFS), a glue providing the elasticity capability for storage resources by federating on demand any POSIX file systems. Although our first prototype is under heavy development, we discuss in this paper the GBFS model and few preliminary but promising results.
Type de document :
Communication dans un congrès
WPBA Workshop in Proceedings of 26th International Symposium on Computer Architecture and High Performance Computing, Oct 2014, Paris, France. IEEE, Workshop on Parallel and Distributed Computing for Big Data Applications - WPBA 2014 (collocated with IEEE SBAC-PAD conference), 2014, WPBA Workshop in Proceedings of 26th International Symposium on Computer Architecture and High Performance Computing. 〈http://wpba2014.dc.ufscar.br/index.html〉
Liste complète des métadonnées

https://hal.inria.fr/hal-01085233
Contributeur : Adrien Lebre <>
Soumis le : jeudi 20 novembre 2014 - 21:39:39
Dernière modification le : mercredi 11 avril 2018 - 02:01:23

Identifiants

  • HAL Id : hal-01085233, version 1

Citation

Adrien Lèbre, Gustavo Bervian Brand. GBFS: Efficient Data-Sharing on Hybrid Platforms. Towards adding WAN-Wide elasticity to DFSes.. WPBA Workshop in Proceedings of 26th International Symposium on Computer Architecture and High Performance Computing, Oct 2014, Paris, France. IEEE, Workshop on Parallel and Distributed Computing for Big Data Applications - WPBA 2014 (collocated with IEEE SBAC-PAD conference), 2014, WPBA Workshop in Proceedings of 26th International Symposium on Computer Architecture and High Performance Computing. 〈http://wpba2014.dc.ufscar.br/index.html〉. 〈hal-01085233〉

Partager

Métriques

Consultations de la notice

183