ReSolve is a library of GPU-resident linear solvers. It contains iterative and direct solvers designed to run on NVIDIA and AMD GPUs, as well as on CPU devices.
The User Guide and developer's documentation is available online, including Doxygen-generated source code documentation.
Dependencies:
- KLU, AMD and COLAMD libraries from SuiteSparse >= 5.0
- CMake >= 3.22
- CUDA >= 11.4 (optional)
- HIP/ROCm >= 5.6 (optional)
To build it:
$ git clone https://github.com/ORNL/ReSolve.git
$ mkdir build && cd build
$ cmake ../ReSolve
$ makeIn the directory where you built the library run
$ make installTo change the install location please use the CMAkePresets.json file as mentioned in test and deploy
To run it, download test linear systems and then edit script runResolve to match locations of your linear systems and binary installation. The script will emulate nonlinear solver calling the linear solver repeatedly.
Make sure Resolve library is installed (see above)
Below is an example CMakeList.txt file to use ReSolve library in your project
cmake_minimum_required(VERSION 3.20)
project(my_app LANGUAGES CXX)
find_package(ReSolve CONFIG
PATHS ${ReSolve_DIR} ${ReSolve_DIR}/share/resolve/cmake
ENV LD_LIBRARY_PATH ENV DYLD_LIBRARY_PATH
REQUIRED)
# Build your executable
add_executable(my_app my_app.cpp)
target_link_libraries(my_app PRIVATE ReSolve::ReSolve)For all contributions to ReSolve please follow the developer guidelines
ReSolve as a library is tested every merge request via Gitlab pipelines that execute various library tests including a test of ReSolve being consumed as package within an external project as mentioned in Using ReSolve in your own Project
To test your own install of ReSolve simply run from your ReSolve build directory
$ make testAfter you make install you can test your installation by running
$ make test_installfrom your build directory.
You can find default Cmake Configurations in the CMakePresets.json file, which allows for easy switching between different CMake Configs. To create your own CMake Configuration we encourage you to utlize a CmakeUserPresets.json file. To learn more about cmake-presets please checkout the cmake docs
For example if you wanted to build and install ReSolve on a High Performance Computing Cluster such as PNNL's Deception or ORNL's Ascent we encourage you to utilize our cluster preset. Using this preset will set CMAKE_INSTALL_PREFIX to an install folder. To use this preset simply call the preset flag in the cmake build step.
cmake -B build --preset clusterFor technical questions or to report a bug please submit a GitHub issue. For non-technical issues please contact Kasia Świrydowicz kasia.swirydowicz@pnnl.gov or Slaven Peles peless@ornl.gov.
Primary authors of this project are Kasia Świrydowicz and Slaven Peles.
ReSolve project would not be possible without significant contributions from (in alphabetic order):
- Maksudul Alam
- Ryan Danehy
- Nicholson Koukpaizan
- Jaelyn Litzinger
- Phil Roth
- Cameron Rutherford
Development of this code was supported by the Exascale Computing Project (ECP), Project Number: 17-SC-20-SC, a collaborative effort of two DOE organizations—the Office of Science and the National Nuclear Security Administration—responsible for the planning and preparation of a capable exascale ecosystem—including software, applications, hardware, advanced system engineering, and early testbed platforms—to support the nation's exascale computing imperative.
Copyright © 2023, UT-Battelle, LLC, and Battelle Memorial Institute.
ReSolve is a free software distributed under a BSD-style license. See the LICENSE and NOTICE files for details. All new contributions to ReSolve must be made under the smae licensing terms.
Please Note If you are using ReSolve with any third party libraries linked in (e.g., KLU), be sure to review the respective license of the package as that license may have more restrictive terms than the ReSolve license.
