Investigating Portability in Chapel for Tree-based Optimization on GPU-powered Clusters

Carneiro, Tiago; Kayraklioglu, Engin; Helbecque, Guillaume; Melab, Nouredine

Type de document :

Communication dans un congrès avec actes

DOI :

10.1007/978-3-031-69583-4_27

Titre :

Investigating Portability in Chapel for Tree-based Optimization on GPU-powered Clusters

Auteur(s) :

Carneiro, Tiago [Auteur]
IMEC [IMEC]
Kayraklioglu, Engin [Auteur]
Hewlett Packard Enterprise (Hewlett Packard) [HPE]
Helbecque, Guillaume [Auteur]
Optimisation de grande taille et calcul large échelle [BONUS]
Inria Lille - Nord Europe
Centre de Recherche en Informatique, Signal et Automatique de Lille - UMR 9189 [CRIStAL]
Université de Lille
Université du Luxembourg = University of Luxembourg = Universität Luxemburg [uni.lu]
Melab, Nouredine [Auteur]
Optimisation de grande taille et calcul large échelle [BONUS]
Inria Lille - Nord Europe
Centre de Recherche en Informatique, Signal et Automatique de Lille - UMR 9189 [CRIStAL]
Université de Lille

Titre de la manifestation scientifique :

30th International European Conference on Parallel and Distributed Computing

Ville :

Madrid

Pays :

Espagne

Date de début de la manifestation scientifique :

2024-08-26

Titre de l’ouvrage :

Euro-Par 2024: Parallel Processing

Lieu de publication :

Cham

Date de publication :

2024-08-26

Mot(s)-clé(s) en anglais :

GPU Computing
Backtracking
Chapel
Portability

Discipline(s) HAL :

Informatique [cs]/Calcul parallèle, distribué et partagé [cs.DC]

Résumé en anglais : [en]

The Top500 list features supercomputers powered by accelerators from different vendors. This variety brings, along with the heterogeneity challenge, both the code and performance portability challenges. In this context, ...
Lire la suite >The Top500 list features supercomputers powered by accelerators from different vendors. This variety brings, along with the heterogeneity challenge, both the code and performance portability challenges. In this context, Chapel's native GPU support comes as a solution for code portability between different vendors. In this paper, we investigate the viability of using the Chapel high-productivity language as a tool to achieve both code and performance portability in large-scale tree-based search. As a case study, we implemented a distributed backtracking for solving permutation combinatorial problems. Extensive experiments conducted on big N-Queens problem instances, using up to 512 NVIDIA GPUs and 1024 AMD GPUs on Top500 supercomputers, reveal that it is possible to scale on the two different systems using the same tree-based search written in Chapel. This trade-off results in a performance decrease of less than 10% for the biggest problem instances.Lire moins >

Langue :

Anglais

Comité de lecture :

Oui

Audience :

Internationale

Vulgarisation :

Non

Projet ANR :

Calcul ultra-scale pour la résolution de problèmes d'optimisation de grande taille

Collections :

Centre de Recherche en Informatique, Signal et Automatique de Lille (CRIStAL) - UMR 9189

Source :

Harvested from HAL

Investigating Portability in Chapel for ... BibTeX CSV Excel RIS

Investigating Portability in Chapel for ...

BibTeX

CSV

Excel

RIS