In this work, we present our port of OpenFOAM to GPUs using the C++ standard parallel execution model (stdpar) introduced in ISO C++17. With a minimally intrusive strategy—primarily replacing serial loops with stdpar constructs—we offload the full computational workload of typical CFD simulations to multicore and manycore architectures. This approach is vendor\u2011agnostic, maintains a single unified codebase, and can be integrated easily into the main OpenFOAM release.
\nWe demonstrate performance using the icoFoam and simpleFoam solvers across four test cases: the 3D lid\u2011driven cavity, 3D conical diffuser, HPC motorbike, and drivAer automotive geometry. Experiments were conducted on a range of NVIDIA and AMD systems, including CPU\u2011only, hybrid CPU–GPU, and unified\u2011memory CPU–GPU configurations. Measured speedups relative to a fully populated 32\u2011core CPU socket range from 0.4× to 7.7×, depending on boundary\u2011condition complexity, turbulence modelling, and solver type. Details of the porting methodology and performance results are provided.
", "languages": [ { "id": "eng", "title": { "en": "English" } } ], "publication_date": "2026-03-19", "publisher": "Zenodo", "resource_type": { "id": "poster", "title": { "de": "Poster", "en": "Poster" } }, "rights": [ { "description": { "en": "" }, "icon": "cc-by-nc-nd-icon", "id": "cc-by-nc-nd-4.0", "props": { "scheme": "spdx", "url": "https://creativecommons.org/licenses/by-nc-nd/4.0/legalcode" }, "title": { "en": "Creative Commons Attribution Non Commercial No Derivatives 4.0 International" } } ], "subjects": [ { "subject": "Standard Parallelism C++" }, { "subject": "CFD" }, { "subject": "OpenFOAM" }, { "subject": "ARCHER2" }, { "subject": "eCSE" }, { "subject": "Celebration of Science" } ], "title": "Cross-platform Programming Model for GPU Implementation of OpenFOAM Using ISO C++", "version": "1.0" }, "parent": { "access": { "owned_by": { "user": "48291" }, "settings": { "accept_conditions_text": null, "allow_guest_requests": false, "allow_user_requests": false, "secret_link_expiration": 0 } }, "communities": { "default": "ac3b180d-8b5d-4f22-84a9-35adfb25eed3", "entries": [ { "access": { "member_policy": "open", "members_visibility": "public", "record_submission_policy": "open", "review_policy": "open", "visibility": "public" }, "children": { "allow": false }, "created": "2021-03-10T16:01:47.749254+00:00", "custom_fields": {}, "deletion_status": { "is_deleted": false, "status": "P" }, "id": "ac3b180d-8b5d-4f22-84a9-35adfb25eed3", "links": {}, "metadata": { "curation_policy": "This community is to share the open access materials relevant to ARCHER2 only.
", "description": "ARCHER2 is the UK National Supercomputing Service. This is the ARCHER2 community on Zenodo repository.", "page": "The ARCHER2 Service is a world class advanced computing resource for UK researchers. ARCHER2 is provided by UKRI, EPCC, HPE Cray and the University of Edinburgh. ARCHER2 is currently running as a small early access system while the full system is being commissioned.
\n\nThis ARCHER2 community on zenodo is to share open access materials relevant to ARCHER2, such as ARCHER2 white papers, training materials, publications, eCSE technical reports, etc. https://www.archer2.ac.uk
", "title": "ARCHER2", "website": "https://www.archer2.ac.uk" }, "revision_id": 1, "slug": "archer2", "updated": "2024-03-21T08:59:15.656090+00:00" } ], "ids": [ "ac3b180d-8b5d-4f22-84a9-35adfb25eed3" ] }, "id": "19257542", "pids": { "doi": { "client": "datacite", "identifier": "10.5281/zenodo.19257542", "provider": "datacite" } } }, "pids": { "doi": { "client": "datacite", "identifier": "10.5281/zenodo.19257543", "provider": "datacite" }, "oai": { "identifier": "oai:zenodo.org:19257543", "provider": "oai" } }, "revision_id": 4, "stats": { "all_versions": { "data_volume": 19789172.0, "downloads": 13, "unique_downloads": 12, "unique_views": 19, "views": 19 }, "this_version": { "data_volume": 19789172.0, "downloads": 13, "unique_downloads": 12, "unique_views": 19, "views": 19 } }, "status": "published", "swh": {}, "updated": "2026-03-27T15:16:32.695661+00:00", "versions": { "index": 1, "is_latest": true } }