Compact-Kernel (CK) MPM

This repository contains the official implementation of the Compact-Kernel Material Point Method (CK-MPM) and a set of example tests. The tests are built as standalone executables using CMake and require a working CUDA toolchain. The code is built and tested on Ubuntu.

Prerequisites

• CMake 3.27 or newer
• A C++20 compatible compiler with CUDA support (tested with the NVIDIA CUDA Toolkit)
• libglut-dev
• python3-dev
• Git & Git-Lfs

Cloning the Repository

This project uses Git submodules for its third‑party dependencies. Make sure to clone the repository recursively:

git clone --recursive https://github.com/AppledoreM/CKMPM 

If you already cloned the repository without --recursive, initialise the submodules manually:

git submodule update --init --recursive

Building

Create a build directory and invoke CMake. All targets can then be compiled via cmake --build or your favourite build tool:

mkdir build
cd build
cmake ..
cmake --build . -j8

This will produce several test executables under build/tests/. Each executable is named mpm_test_<name> or similar, depending on the test case defined in tests/CMakeLists.txt.

Running the Tests

Run the desired test executable from the build directory. For example:

./tests/mpm_test_dragon

Before running any tests, create a result directory in the repository root to store output files:

mkdir -p result

Some tests export their results to result/<test_name>/. Tests are executed individually.

Python Implementation

This repo also contains a python implementation of PIC version of CK-MPM, which is included in the python-src folder. Check out the .sh file on how to run the script.

License

This project is licensed under the Apache 2.0 License - see the LICENSE file for details.

Citation

@article{liu2025ckmpm,
  author = {Michael Liu, Xinlei Wang, Minchen Li},
  title = {CK-MPM: A Compact-Kernel Material Point Method},
  year = {2025},
  issue_date = {July 2025},
  publisher = {Association for Computing Machinery},
  address = {New York, NY, USA},
  volume = {44},
  number = {4},
  issn = {0730-0301},
  url = {https://doi.org/10.1145/3731155},
  doi = {10.1145/3731155},
  journal = {ACM Trans. Graph.},
  month = july,
  numpages = {14}
}