UHAINA : A parallel high performance unstructured near-shore wave model - Inria - Institut national de recherche en sciences et technologies du numérique Access content directly
Conference Papers Year : 2018

UHAINA : A parallel high performance unstructured near-shore wave model

Abstract

UHAINA is a new phase-resolving free surface wave model for coastal engineering 21 problems. It is based on the most advanced and recent contributions in coastal 22 modelling from the french institutes EPOC, IMAG, IMB, and INRIA BSO. It solves a 23 non-classical version of the depth-integrated fully-nonlinear and weakly-dispersive 24 equations of Green-Naghdi, which allows an efficient numerical implementation. 25 UHAINA relies on libraries developed at the INRIA BSO center, such as AeroSol for 26 its hydrodynamic core, and PaMPA and SCOTCH to handle data management for 27 distributed memory parallel computation. The use of these libraries, in particular 28 AeroSol, offers a wide range of possibilities including arbitrary high-order finite 29 element discretizations, hybrid meshes (structured and unstructured), as well as an 30 advanced programming environment specially designed by the purpose of performance 31 and HPC. These properties will lead in the coming years to the release of a new efficient 32 and robust open source wave modelling platform, available for a large community of 33 users and very suitable for practical coastal applications.
Fichier principal
Vignette du fichier
UHAINA_A_parallel_high_performance_unstructured_ad.pdf (396.22 Ko) Télécharger le fichier
Origin : Files produced by the author(s)
Loading...

Dates and versions

hal-01824108 , version 1 (26-06-2018)

Identifiers

Cite

Andrea Gilberto Filippini, Sébastien de Brye, Vincent Perrier, Fabien Marche, Mario Ricchiuto, et al.. UHAINA : A parallel high performance unstructured near-shore wave model. Journées Nationales Génie Côtier - Génie Civil, May 2018, La Rochelle, France. pp.47-56, ⟨10.5150/jngcgc.2018.006⟩. ⟨hal-01824108⟩
1311 View
495 Download

Altmetric

Share

Gmail Facebook X LinkedIn More