Fast Biological Sequence Comparison on Hybrid Platforms

Abstract : Today, many high performance computing plat-forms use hybrid architectures combining multi-core processors and hardware accelerators like GPUs (Graphic Processing Units). This paper presents a new method for scheduling tasks for biological sequence comparison applications with CPUs and GPUs. This strategy is called SWDUAL and is based on a dual approximation scheme for determining which tasks are most suitable to be executed on the GPUs. The objective is to obtain fast execution time and minimize the idle time on each PE (Processing Element). It is implemented using a master-slave model. Results obtained when sequences were compared to five public genomic databases show that this method allows to reduce the execution time on hybrid platforms when compared to other public available implementations.
Type de document :
Communication dans un congrès
43rd International Conference on Parallel Processing, ICPP 2014, Sep 2014, Minneapolis, United States. pp.501 - 509, 2014, 〈10.1109/ICPP.2014.59〉
Liste complète des métadonnées

Littérature citée [19 références]  Voir  Masquer  Télécharger

https://hal.archives-ouvertes.fr/hal-01102263
Contributeur : Grégory Mounié <>
Soumis le : lundi 12 janvier 2015 - 13:56:17
Dernière modification le : vendredi 21 décembre 2018 - 10:46:08
Document(s) archivé(s) le : samedi 12 septembre 2015 - 01:30:18

Fichier

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

Identifiants

Collections

Citation

Safia Kedad-Sidhoum, Fernando Mendonca, Florence Monna, Gregory Mounié, Denis Trystram. Fast Biological Sequence Comparison on Hybrid Platforms. 43rd International Conference on Parallel Processing, ICPP 2014, Sep 2014, Minneapolis, United States. pp.501 - 509, 2014, 〈10.1109/ICPP.2014.59〉. 〈hal-01102263〉

Partager

Métriques

Consultations de la notice

477

Téléchargements de fichiers

122