Skip to content

TNL-Project/tnl-benchmark-mesh

Repository files navigation

tnl-benchmark-mesh

Benchmarks for the unstructured mesh in TNL. Two algorithms (calculation of cell measures and calculation of curve lengths/surface areas around vertices) were presented in the paper Configurable open-source data structure for distributed conforming unstructured homogeneous meshes with GPU support.

Quickstart

  1. Make sure that Git, Git LFS, CMake, GNU Make, Python 3, and CUDA 11 are installed on your Linux system.

  2. Clone the repository:

    git clone https://gitlab.com/tnl-project/tnl-benchmark-mesh.git
    
  3. Build the benchmark binaries:

    cd tnl-benchmark-mesh
    python3 MeshBenchmarksRunner.py
    cmake -B build -S . -G "Unix Makefiles" -DCMAKE_BUILD_TYPE=Release
    cmake --build build
    
  4. Run the script to execute the benchmarks on the meshes included in the repository:

    ./run_benchmark
    

About

Benchmarks for the unstructured mesh in TNL.

Topics

Resources

License

Stars

Watchers

Forks

Releases

No releases published

Packages

 
 
 

Contributors