Efficient sweep kernels on shared-memory architectures for the discrete ordinates neutron transport equation on Cartesian and hexagonal geometries
Year: 2024
Type: article
Abstract: \nThis paper describes the implementation of DONUT, a small multi-group SN-DG transport solver that aims at providing efficient and portable sweep kernels on shared-memory architectures for Cartesian and hexagonal geometries. DONUT heavily relies on the Kokkos C++ library for portability and genericity. First encouraging performance results are presented for multicore CPU architectures.\n
Authors: Gabriel Suau, Ansar Calloo, Rémi Baron, Romain Le Tellier, Thierry Gautier
Institutions: Commissariat à l'Énergie Atomique et aux Énergies Alternatives, Université Paris-Saclay, CEA Paris-Saclay, Service d'Étude des Réacteurs et de Mathématiques Appliquées, Service de Génie Logiciel pour la Simulation, CEA Cadarache, Institut de Recherche sur les Systèmes Nucléaires pour la production d'Energie bas carbone, École Normale Supérieure de Lyon, Institut national de recherche en sciences et technologies du numérique, Laboratoire de l'Informatique du Parallélisme, AVALON: Algorithms and Software Architectures for Distributed and HPC Platforms
Language: English
Subfield: Aerospace Engineering
Field: Engineering
Domain: Physical Sciences
Open Access status: green
OpenAlex ID: W6903204149
FWCI:
Cites:
Cited by:
Related to: