SAMRAI (Structured Adaptive Mesh Refinement Application Infrastructure) is an object-oriented C++ software library that enables exploration of numerical, algorithmic, parallel computing, and software issues associated with applying structured adaptive mesh refinement (SAMR) technology in large-scale parallel application development. SAMRAI provides software tools for developing SAMR applications that involve coupled physics models, sophisticated numerical solution methods, and which require high-performance parallel computing hardware. SAMRAI enables integration of SAMR technology into existing codes and simplifies the exploration of SAMR methods in new application domains.
The current release is SAMRAI v. 4.0.1. With the version 4 release, the SAMRAI project is pleased to introduce new features that support running applications on GPU-based architectures, using capabilities provided by the Umpire and RAJA libraries.
SAMRAI is an open source project, and questions, discussion and contributions are welcome!
To get in touch with all the SAMRAI developers, please email samrai@llnl.gov
Contributing to SAMRAI should be easy! We are managing contributions through
pull requents here on GitHub. When you create your pull request, please make
master the target branch.
Your PR must pass all of SAMRAI's unit tests, which are enforced using Travis CI. For information on how to run these tests locally, please see our contribution guidelines
The master branch contains the latest development, and releases are tagged.
New features should be created in feature/<name>branches and be based on
master.
We maintain a list of publications here.
Copyright (c) 1997-2024, Lawrence Livermore National Security, LLC.
Produced at the Lawrence Livermore National Laboratory.
All rights reserved.
Released under LGPL v2.1
For release details and restrictions, please read the LICENSE file. It is also linked here: LICENSE
LLNL-CODE-434871