Since 1993, the CEA has developed TRUST, an open-source CFD software platform designed to address a wide range of thermohydraulic problems. Initially focused on nuclear applications, the platform has progressively evolved to support incompressible single-phase flows, low-Mach-number reactive flows, and fully compressible multi-phase flows. TRUST incorporates a variety of numerical schemes and supports multiple mesh types, making it suitable for both academic and industrial applications, as well as for deployment on standard computing systems and high-performance computing (HPC) clusters. Notably, it has enabled one of the largest Direct Numerical Simulations (DNS) to date for hydrogen risk assessment, involving over 2 billion cells and more than 50,000 MPI processors. Recently, the development team has begun integrating GPU-based computing technologies\u2013such as AmgX, rocALUTION, AMGCL and Kokkos\u2013to develop a hybrid CPU\/GPU version of the code capable of delivering high performance across diverse architectures. This article presents an overview of the TRUST platform, reviews the largest simulation performed to date with the CPU version, outlines the ongoing GPU integration strategy, and highlights selected results.<\/p>\n","protected":false},"excerpt":{"rendered":"
Since 1993, the CEA has developed TRUST, an open-source CFD software platform designed to address a wide range of thermohydraulic problems. Initially focused on nuclear applications, the platform has progressively evolved to support incompressible single-phase flows, low-Mach-number reactive flows, and fully compressible multi-phase flows. TRUST incorporates a variety of numerical schemes and supports multiple mesh […]<\/p>\n","protected":false},"author":351,"featured_media":0,"comment_status":"open","ping_status":"closed","sticky":false,"template":"","format":"standard","meta":{"_jetpack_memberships_contains_paid_content":false,"footnotes":"","jetpack_publicize_message":"","jetpack_publicize_feature_enabled":true,"jetpack_social_post_already_shared":true,"jetpack_social_options":{"image_generator_settings":{"template":"highway","default_image_id":0,"font":"","enabled":false},"version":2}},"categories":[89,104,3],"tags":[2189,2150,7,1600,14,1795,2063,242,285,20,2066,252,176],"class_list":["post-30266","post","type-post","status-publish","format-standard","hentry","category-nvidia-cuda","category-fluid-dynamics","category-paper","tag-amd-radeon-instinct-mi250a","tag-amd-radeon-instinct-mi300a","tag-ati","tag-cfd","tag-cuda","tag-fluid-dynamics","tag-hip","tag-mpi","tag-numerical-simulation","tag-nvidia","tag-nvidia-a100","tag-openmp","tag-package"],"views":1854,"jetpack_publicize_connections":[],"jetpack_featured_media_url":"","jetpack_sharing_enabled":true,"_links":{"self":[{"href":"https:\/\/hgpu.org\/index.php?rest_route=\/wp\/v2\/posts\/30266","targetHints":{"allow":["GET"]}}],"collection":[{"href":"https:\/\/hgpu.org\/index.php?rest_route=\/wp\/v2\/posts"}],"about":[{"href":"https:\/\/hgpu.org\/index.php?rest_route=\/wp\/v2\/types\/post"}],"author":[{"embeddable":true,"href":"https:\/\/hgpu.org\/index.php?rest_route=\/wp\/v2\/users\/351"}],"replies":[{"embeddable":true,"href":"https:\/\/hgpu.org\/index.php?rest_route=%2Fwp%2Fv2%2Fcomments&post=30266"}],"version-history":[{"count":0,"href":"https:\/\/hgpu.org\/index.php?rest_route=\/wp\/v2\/posts\/30266\/revisions"}],"wp:attachment":[{"href":"https:\/\/hgpu.org\/index.php?rest_route=%2Fwp%2Fv2%2Fmedia&parent=30266"}],"wp:term":[{"taxonomy":"category","embeddable":true,"href":"https:\/\/hgpu.org\/index.php?rest_route=%2Fwp%2Fv2%2Fcategories&post=30266"},{"taxonomy":"post_tag","embeddable":true,"href":"https:\/\/hgpu.org\/index.php?rest_route=%2Fwp%2Fv2%2Ftags&post=30266"}],"curies":[{"name":"wp","href":"https:\/\/api.w.org\/{rel}","templated":true}]}}