This directory contains a set of scripts designed to benchmark NetworkX with the nx-cugraph backend and deliver a report that summarizes the speed-up and runtime deltas over default NetworkX.

Our current benchmarks provide the following datasets:

Install nx-cugraph and the benchmark dependencies:

conda create -n nx-cugraph-benchmark -c rapidsai -c conda-forge nx-cugraph pytest pytest-benchmark

This script allows users to run a small set of commonly-used algorithms across multiple datasets and backends. All results are stored inside a sub-directory (logs/) and output files are named based on the combination of parameters for that benchmark.

NOTE: If running with all algorithms and datasets using NetworkX without an accelerated backend, this script may take a few hours to finish running.

Usage:

• Run with --cpu-only:

./run-main-benchmarks.sh --cpu-only

• Run with --gpu-only:

./run-main-benchmarks.sh --gpu-only

• Run without any arguments (all backends):

./run-main-benchmarks.sh

This script downloads the specified dataset using cugraph.datasets.

Usage:

python get_graph_bench_dataset.py [dataset]

This script is designed to be run after run-gap-benchmarks.sh in order to generate an HTML page displaying a results table comparing default NetworkX to nx-cugraph. The script also provides information about the current system, so it should be run on the machine on which benchmarks were run.

Usage:

python create_results_summary_page.py > report.html