Geodesic
A Task-Driven Implementation of a Simple Numerical Solver for Hyperbolic Conservation Laws
Mohamed Essadki1, 2 ; Jonathan Jung3, 4 ; Adam Larat1, 5 ; Milan Pelletier1 ; Vincent Perrier4, 3
1Laboratoire EM2C, CNRS, CentraleSupélec, Université Paris Saclay,, Grande Voie des Vignes,Châtenay-Malabry, 92295, France
2IFP Énergies nouvelles, 1-4 avenue de Bois-Préau,Rueil-Malmaison Cedex, 92852, France
3LMAP UMR 5142, UPPA, CNRS
4Cagire Team, INRIA Bordeaux Sud-Ouest, Pau- France
5Fédération de Mathématiques de l’École Centrale Paris, CNRS FR 3487, Grande Voie des Vignes,Châtenay-Malabry, 92295
ESAIM. Proceedings, Tome 63 (2018), pp. 228-247
Cet article a éte moissonné depuis la source EDP Sciences

Voir la notice de l'article

This article describes the implementation of an all-in-one numerical procedure within the runtime StarPU. In order to limit the complexity of the method, for the sake of clarity of the presentation of the non-classical task-driven programming environment, we have limited the numerics to first order in space and time. Results show that the task distribution is efficient if the tasks are numerous and individually large enough so that the task heap can be saturated by tasks which computational time covers the task management overhead. Next, we also see that even though they are mostly faster on graphic cards, not all the tasks are suitable for GPUs, which brings forward the importance of the task scheduler. Finally, we look at a more realistic system of conservation laws with an expensive source term, what allows us to conclude and open on future works involving higher local arithmetic intensity, by increasing the order of the numerical method or by enriching the model (increased number of parameters and therefore equations).
DOI : 10.1051/proc/201863228

Mohamed Essadki  1 , 2   ; Jonathan Jung  3 , 4   ; Adam Larat  1 , 5   ; Milan Pelletier  1   ; Vincent Perrier  4 , 3

1 Laboratoire EM2C, CNRS, CentraleSupélec, Université Paris Saclay,, Grande Voie des Vignes,Châtenay-Malabry, 92295, France
2 IFP Énergies nouvelles, 1-4 avenue de Bois-Préau,Rueil-Malmaison Cedex, 92852, France
3 LMAP UMR 5142, UPPA, CNRS
4 Cagire Team, INRIA Bordeaux Sud-Ouest, Pau- France
5 Fédération de Mathématiques de l’École Centrale Paris, CNRS FR 3487, Grande Voie des Vignes,Châtenay-Malabry, 92295 
@article{EP_2018_63_a10,
     author = {Mohamed Essadki and Jonathan Jung and Adam Larat and Milan Pelletier and Vincent Perrier},
     title = {A {Task-Driven} {Implementation} of a {Simple} {Numerical} {Solver} for {Hyperbolic} {Conservation} {Laws}},
     journal = {ESAIM. Proceedings},
     pages = {228--247},
     year = {2018},
     volume = {63},
     doi = {10.1051/proc/201863228},
     language = {en},
     url = {http://geodesic.mathdoc.fr/articles/10.1051/proc/201863228/}
}
TY  - JOUR
AU  - Mohamed Essadki
AU  - Jonathan Jung
AU  - Adam Larat
AU  - Milan Pelletier
AU  - Vincent Perrier
TI  - A Task-Driven Implementation of a Simple Numerical Solver for Hyperbolic Conservation Laws
JO  - ESAIM. Proceedings
PY  - 2018
SP  - 228
EP  - 247
VL  - 63
UR  - http://geodesic.mathdoc.fr/articles/10.1051/proc/201863228/
DO  - 10.1051/proc/201863228
LA  - en
ID  - EP_2018_63_a10
ER  - 
%0 Journal Article
%A Mohamed Essadki
%A Jonathan Jung
%A Adam Larat
%A Milan Pelletier
%A Vincent Perrier
%T A Task-Driven Implementation of a Simple Numerical Solver for Hyperbolic Conservation Laws
%J ESAIM. Proceedings
%D 2018
%P 228-247
%V 63
%U http://geodesic.mathdoc.fr/articles/10.1051/proc/201863228/
%R 10.1051/proc/201863228
%G en
%F EP_2018_63_a10
Mohamed Essadki; Jonathan Jung; Adam Larat; Milan Pelletier; Vincent Perrier. A Task-Driven Implementation of a Simple Numerical Solver for Hyperbolic Conservation Laws. ESAIM. Proceedings, Tome 63 (2018), pp. 228-247. doi: 10.1051/proc/201863228

Cité par Sources :