hipGRAPH

Caution

This release is an early-access software technology preview. Running production workloads is not recommended.

Note

This ROCm™ port is derived work based on the NVIDIA RAPIDS® cugraph project's C API. The CUDA back-end is not tested regularly.

hipGRAPH is a graph marshalling library that acts as a wrapper between your application and the 'worker' graph library, rocGRAPH <https://github.com/ROCm-DS/rocGraph>. The hipGraph library contains a collection of graph algorithms, enabling you to build, analyze, and manipulate complex graphs or networks. hipGraph is derived from the cuGraph library that forms part of the NVIDIA RAPIDS open source project. The hipGraph library described here should not be confused with HIP graphs <https://rocm.docs.amd.com/projects/HIP/en/latest/how-to/hip_runtime_api/hipgraph.html>_ in the HIP runtime API.

Documentation

Documentation for hipGRAPH is in progress.

Build hipGRAPH

To build hipGRAPH, you can use our bash helper script (for Ubuntu, Centos, RHEL, Fedora, SLES, openSUSE-Leap) or you can perform a manual build (for all supported platforms).

• Bash helper script (install.sh): This script, which is located in the root of this repository, builds and installs hipGRAPH on Ubuntu with a single command. Note that this option doesn't allow much customization and hard-codes configurations that can be specified through invoking CMake directly. Some commands in the script require sudo access, so it may prompt you for a password.

  ./install.sh -h  # shows help
  ./install.sh -id # builds library, dependencies, then installs (the `-d` flag only needs to be passed once on a system)

• Manual build: If you use a distribution other than Ubuntu, or would like more control over the build process, run cmake manually. See the install

  mkdir build
  cd build
  cmake .. -DCMAKE_CXX_COMPILER="${rocm_path}/bin/hipcc" \
      -DCMAKE_C_COMPILER="${rocm_path}/bin/hipcc" \
      -DCMAKE_BUILD_TYPE=Debug
  # Add -DUSE_CUDA=ON to enable the NVIDIA back-end

Python modules

The modules overall are a work in progress. pylibhipgraph is functional and can be build from the python subdirectory with

python3 -m build pylibhipgraph/

or

conda run -n dev python3 -m build pylibhipgraph/

once all the prerequisites are installed. Documentation for installing the prerequisites is in progress.