Getting started with cuGraph

Required hardware/software

CuGraph is part of Rapids and has the following system requirements:

• NVIDIA GPU, Volta architecture or later, with compute capability 7.0+

• CUDA 11.2, 11.4, 11.5, 11.8, 12.0, 12.2, or 12.5

• Python version 3.10, 3.11, or 3.12

• NetworkX >= version 3.3 or newer in order to use use NetworkX Configs This is required for use of nx-cuGraph, see below.

Installation

The latest RAPIDS System Requirements documentation is located here.

This includes several ways to set up cuGraph

• From Unix

  • Conda

  • Docker

  • pip

• In windows you must install WSL2 and then choose one of the following:

  • Conda

  • Docker

  • pip

• Build From Source

To build from source, check each RAPIDS GitHub README for set up and build instructions. Further links are provided in the selector tool. If additional help is needed reach out on our Slack Channel.

CuGraph Using NetworkX Code

While the steps above are required to use the full suite of cuGraph graph analytics, cuGraph is now supported as a NetworkX backend using nx-cugraph. Nx-cugraph offers those with existing NetworkX code, a zero code change option with a growing list of supported algorithms.

Cugraph API Example

Coming soon !

Until then, the cuGraph notebook repository has many examples of loading graph data and running algorithms in Jupyter notebooks. The cuGraph test code gives examples of python scripts settng up and calling cuGraph algorithms. A simple example of testing the degree centrality algorithm is a good place to start. Some of these examples show multi-GPU tests/examples with larger data sets as well.