Issue |
ESAIM: Proc.
Volume 43, December 2013
CEMRACS 2012
|
|
---|---|---|
Page(s) | 59 - 79 | |
DOI | https://doi.org/10.1051/proc/201343004 | |
Published online | 13 December 2013 |
FullSWOF Paral: Comparison of two parallelization strategies (MPI and SKELGIS) on a software designed for hydrology applications*,**
1 MAPMO, fédération Denis Poisson,
University of Orléans, France,
e-mail : stephane.cordier@math.cnrs.fr
2 LIFO, University of Orléans &
Géo-Hyd, France,
e-mail : helene.coullon@univ-orleans.fr
3 Lab. J.A. Dieudonné & EPU Nice
Sophia, University of
Nice,
France,
e-mail : delestre@math.unice.fr
4 MAPMO, fédération Denis Poisson,
University of
Orléans,
France,
e-mail : christian.laguerre@math.cnrs.fr
5 MAPMO & BRGM, University
of
Orléans,
France,
e-mail : Minh.Hoang.Le@math.cnrs.fr
6 Géo-Hyd, 101 rue Jacques Charles, 45160
Olivet,
France,
e-mail : daniel.pierre@geo-hyd.com
7 LAMFA, University of
Picardie, France,
e-mail : georges.sadaka@u-picardie.fr
In this paper, we perform a comparison of two approaches for the parallelization of an existing, free software, FullSWOF2D (http ://www.univ-orleans.fr/mapmo/soft/FullSWOF/ that solves shallow water equations for applications in hydrology) based on a domain decomposition strategy. The first approach is based on the classical MPI library while the second approach uses Parallel Algorithmic Skeletons and more precisely a library named SkelGIS (Skeletons for Geographical Information Systems). The first results presented in this article show that the two approaches are similar in terms of performance and scalability. The two implementation strategies are however very different and we discuss the advantages of each one.
© EDP Sciences, SMAI 2013
Current usage metrics show cumulative count of Article Views (full-text article views including HTML views, PDF and ePub downloads, according to the available data) and Abstracts Views on Vision4Press platform.
Data correspond to usage on the plateform after 2015. The current usage metrics is available 48-96 hours after online publication and is updated daily on week days.
Initial download of the metrics may take a while.