Benchmarks
=================

This page presents the performance of algorithms currently supported by ``nx-cugraph`` across four graphs of varying sizes. 

The goal is to provide a clear comparison of how dispatching to a GPU-accelerated backend compares against the default, CPU-based implementation of NetworkX. This allows users to get an idea of potential speedups that they might see by leveraging dispatching to a GPU with ``nx-cugraph``. 

As datasets grow larger, the GPU-accelerated backend typically begins to show increasingly faster speedups over the CPU. This trend demonstrates how the GPU's parallel processing capabilities allow it to handle large-scale graph analytics much more efficiently than the CPU.

Datasets
************

The datasets used in the benchmarks are:

.. raw:: html

  <style>
  .data-table {
    width: 95%;
    margin: 0 auto;
    font-size: 15px;
  }
  table tr:hover td {
    text-decoration: underline;
  }
  .data-table th, .data-table td {
    padding: 8px;
    border: 1px solid #ccc;
  }
  .data-table th:nth-child(1), .data-table td:nth-child(1) {
    width: 15%;
    }
  .data-table th:nth-child(2) {
    width: 15%;
  }
  .data-table th:nth-child(3) {
    width: 15%;
  }
  .data-table th:nth-child(4) {
    width: 55%;
  }
  </style>

.. raw:: html

  <table class="data-table">
    <thead>
      <tr>
        <th>Dataset</th>
        <th>Number of Nodes</th>
        <th>Number of Edges</th>
        <th>Description</th>
      </tr>
    </thead>
    <tbody>
      <tr>
        <td><a href="https://sparse.tamu.edu/Newman/netscience">netscience</a></td>
        <td>1,461</td>
        <td>5,484</td>
        <td>A small, directed graph representing a scientific collaboration network.</td>
      </tr>
      <tr>
        <td><a href="https://snap.stanford.edu/data/amazon0302.html">amazon0302</a></td>
        <td>262,111</td>
        <td>1,234,877</td>
        <td>A medium, undirected graph modeling product co-purchasing data.</td>
      </tr>
      <tr>
        <td><a href="https://snap.stanford.edu/data/cit-Patents.html">cit-Patents</a></td>
        <td>3,774,768</td>
        <td>16,518,948</td>
        <td>A large, directed graph representing U.S. patent citation relationships.</td>
      </tr>
      <tr>
        <td><a href="https://snap.stanford.edu/data/soc-LiveJournal1.html">soc-LiveJournal1</a></td>
        <td>4,847,571</td>
        <td>68,993,773</td>
        <td>A very large, directed graph based on the LiveJournal social network.</td>
      </tr>
    </tbody>
  </table>
  </br>

--------------

Algorithm Table
***************

The table below showcases the performance difference between CPU and GPU when running a particular algorithm on one of the four datasets.

Each cell displays the speedup factor achieved when running the algorithm on a ``nx-cugraph`` compared to default NetworkX. A higher value indicates better performance. To enhance readability, the cells are color-coded based on the following speedup criteria:

- Green for excellent (``>5.0``)
- Yellow for moderate (``1.1 - 5.0``)
- Red for suboptimal (``0 - 1.0``)

Also, by hovering over the table, users can see the time it took for NetworkX and nx-cugraph respectively, as well as the delta between both runs. We encourage users to refer to this table to see an approximation of the performance increase they can expect from their workflows.

.. raw:: html

  <style>
  table {
    border-collapse: collapse;
    width: 95%;
    padding: 6px;
    border: 1px solid #ccc;
    font-size: 15px;
  }
  table th, table td {
    padding: 8px;
    text-align: left;
    border: 1px solid rgb(67, 67, 67);
  }
  @media (prefers-color-scheme: dark) {
    td.green, td.red, td.yellow {
        /* dark mode styling */
        color:rgb(30, 30, 30);
    }
  }
  .perf-table th {
    padding: 12px 8px;
  }
  .perf-table td.green {
    background-color: #d4edda;
  }
  .perf-table td.yellow {
    background-color: #fff3cd;
  }
  .perf-table td.red {
    background-color: #f8d7da;
  }
  /* control table width */
  table th:nth-child(1), table td:nth-child(1) {
    width: 40%;
  }
  table :is(th, td):nth-child(n+2):nth-child(-n+5) {
  width: 15%;
  }
  table td {
    position: relative;
  }
  table td .tooltip {
    visibility: hidden;
    position: absolute;
    background-color: rgb(175, 129, 249);
    color: #fff;
    border-radius: 4px;
    padding: 5px;
    font-size: 12px;
    bottom: 100%;
    left: 50%;
    transform: translateX(-50%);
    opacity: 0;
    transition: opacity 0.3s ease;
    white-space: nowrap;
    z-index: 9999;
  }
  table td:hover .tooltip {
    visibility: visible;
    opacity: 1;
  }
  </style>

.. raw:: html

  <table class="perf-table">
    <thead>
      <tr>
        <th>Algorithm</th>
        <th>netscience</th>
        <th>amazon0302</th>
        <th>citpatents</th>
        <th>livejournal</th>
      </tr>
    </thead>
    <tbody>
      <tr>
        <td>ancestors</td>
        <td class="red">0.089*<span class="tooltip">nx: 491.479us<br>nxcg: 5.497ms<br>delta: -5.005ms</span></td>
        <td class="green">60.463<span class="tooltip">nx: 1.712s<br>nxcg: 28.310ms<br>delta: 1.683s</span></td>
        <td class="red">0.134*<span class="tooltip">nx: 4.359ms<br>nxcg: 32.491ms<br>delta: -28.132ms</span></td>
        <td class="green">78.609<span class="tooltip">nx: 56.688s<br>nxcg: 721.145ms<br>delta: 55.967s</span></td>
      </tr>
      <tr>
        <td>average_clustering</td>
        <td class="yellow">1.619*<span class="tooltip">nx: 14.918ms<br>nxcg: 9.213ms<br>delta: 5.706ms</span></td>
        <td class="green">293.432<span class="tooltip">nx: 6.745s<br>nxcg: 22.988ms<br>delta: 6.722s</span></td>
        <td class="green">868.635<span class="tooltip">nx: 181.956s<br>nxcg: 209.474ms<br>delta: 181.747s</span></td>
        <td class="green">2828.128<span class="tooltip">nx: 2151.410s<br>nxcg: 760.719ms<br>delta: 2150.649s</span></td>
      </tr>
      <tr>
        <td>betweenness_centrality, k=10</td>
        <td class="red">0.274*<span class="tooltip">nx: 5.449ms<br>nxcg: 19.923ms<br>delta: -14.474ms</span></td>
        <td class="green">56.98<span class="tooltip">nx: 14.256s<br>nxcg: 250.186ms<br>delta: 14.005s</span></td>
        <td class="green">99.364<span class="tooltip">nx: 95.752s<br>nxcg: 963.653ms<br>delta: 94.789s</span></td>
        <td class="green">264.882<span class="tooltip">nx: 610.395s<br>nxcg: 2.304s<br>delta: 608.091s</span></td>
      </tr>
      <tr>
        <td>betweenness_centrality, k=100</td>
        <td class="red">0.239*<span class="tooltip">nx: 42.450ms<br>nxcg: 177.676ms<br>delta: -135.225ms</span></td>
        <td class="green">50.352<span class="tooltip">nx: 110.614s<br>nxcg: 2.197s<br>delta: 108.417s</span></td>
        <td class="green">520.555<span class="tooltip">nx: 966.550s<br>nxcg: 1.857s<br>delta: 964.694s</span></td>
        <td class="green">487.469<span class="tooltip">nx: 6688.933s<br>nxcg: 13.722s<br>delta: 6675.211s</span></td>
      </tr>
      <tr>
        <td>bfs_edges</td>
        <td class="red">0.035*<span class="tooltip">nx: 295.556us<br>nxcg: 8.488ms<br>delta: -8.192ms</span></td>
        <td class="red">0.383<span class="tooltip">nx: 1.260s<br>nxcg: 3.292s<br>delta: -2.032s</span></td>
        <td class="red">0.003*<span class="tooltip">nx: 115.556us<br>nxcg: 35.910ms<br>delta: -35.794ms</span></td>
        <td class="yellow">2.141<span class="tooltip">nx: 75.646s<br>nxcg: 35.329s<br>delta: 40.318s</span></td>
      </tr>
      <tr>
        <td>bfs_layers</td>
        <td class="red">0.046*<span class="tooltip">nx: 307.147us<br>nxcg: 6.690ms<br>delta: -6.383ms</span></td>
        <td class="green">9.489*<span class="tooltip">nx: 327.838ms<br>nxcg: 34.548ms<br>delta: 293.290ms</span></td>
        <td class="red">0.004*<span class="tooltip">nx: 122.647us<br>nxcg: 32.934ms<br>delta: -32.811ms</span></td>
        <td class="green">33.468<span class="tooltip">nx: 16.313s<br>nxcg: 487.419ms<br>delta: 15.825s</span></td>
      </tr>
      <tr>
        <td>bfs_predecessors</td>
        <td class="red">0.04*<span class="tooltip">nx: 336.403us<br>nxcg: 8.442ms<br>delta: -8.106ms</span></td>
        <td class="red">0.389<span class="tooltip">nx: 1.282s<br>nxcg: 3.293s<br>delta: -2.011s</span></td>
        <td class="red">0.004*<span class="tooltip">nx: 139.603us<br>nxcg: 35.985ms<br>delta: -35.845ms</span></td>
        <td class="yellow">1.728<span class="tooltip">nx: 61.628s<br>nxcg: 35.664s<br>delta: 25.964s</span></td>
      </tr>
      <tr>
        <td>bfs_successors</td>
        <td class="red">0.039*<span class="tooltip">nx: 329.301us<br>nxcg: 8.441ms<br>delta: -8.112ms</span></td>
        <td class="red">0.491<span class="tooltip">nx: 1.802s<br>nxcg: 3.669s<br>delta: -1.866s</span></td>
        <td class="red">0.004*<span class="tooltip">nx: 130.468us<br>nxcg: 35.864ms<br>delta: -35.733ms</span></td>
        <td class="yellow">1.822<span class="tooltip">nx: 62.495s<br>nxcg: 34.297s<br>delta: 28.198s</span></td>
      </tr>
      <tr>
        <td>bfs_tree</td>
        <td class="red">0.121*<span class="tooltip">nx: 855.597us<br>nxcg: 7.060ms<br>delta: -6.204ms</span></td>
        <td class="green">58.569<span class="tooltip">nx: 2.929s<br>nxcg: 50.011ms<br>delta: 2.879s</span></td>
        <td class="red">0.011*<span class="tooltip">nx: 403.871us<br>nxcg: 35.151ms<br>delta: -34.748ms</span></td>
        <td class="green">55.688<span class="tooltip">nx: 74.665s<br>nxcg: 1.341s<br>delta: 73.324s</span></td>
      </tr>
      <tr>
        <td>clustering</td>
        <td class="yellow">1.596*<span class="tooltip">nx: 14.618ms<br>nxcg: 9.162ms<br>delta: 5.456ms</span></td>
        <td class="green">136.185<span class="tooltip">nx: 6.775s<br>nxcg: 49.747ms<br>delta: 6.725s</span></td>
        <td class="green">168.938<span class="tooltip">nx: 181.769s<br>nxcg: 1.076s<br>delta: 180.693s</span></td>
        <td class="green">1132.732<span class="tooltip">nx: 2149.226s<br>nxcg: 1.897s<br>delta: 2147.328s</span></td>
      </tr>
      <tr>
        <td>core_number</td>
        <td class="red">0.223*<span class="tooltip">nx: 2.372ms<br>nxcg: 10.652ms<br>delta: -8.280ms</span></td>
        <td class="green">22.456<span class="tooltip">nx: 1.520s<br>nxcg: 67.706ms<br>delta: 1.453s</span></td>
        <td>--</td>
        <td>--</td>
      </tr>
      <tr>
        <td>degree_centrality</td>
        <td class="red">0.331<span class="tooltip">nx: 361.917us<br>nxcg: 1.092ms<br>delta: -730.266us</span></td>
        <td class="yellow">2.834*<span class="tooltip">nx: 75.040ms<br>nxcg: 26.474ms<br>delta: 48.566ms</span></td>
        <td class="yellow">2.037*<span class="tooltip">nx: 1.565s<br>nxcg: 768.326ms<br>delta: 796.394ms</span></td>
        <td class="yellow">2.045<span class="tooltip">nx: 2.048s<br>nxcg: 1.002s<br>delta: 1.047s</span></td>
      </tr>
      <tr>
        <td>descendants</td>
        <td class="red">0.059*<span class="tooltip">nx: 325.657us<br>nxcg: 5.552ms<br>delta: -5.226ms</span></td>
        <td class="green">38.477<span class="tooltip">nx: 1.259s<br>nxcg: 32.708ms<br>delta: 1.226s</span></td>
        <td class="red">0.004*<span class="tooltip">nx: 127.983us<br>nxcg: 31.525ms<br>delta: -31.397ms</span></td>
        <td class="green">82.354<span class="tooltip">nx: 61.836s<br>nxcg: 750.849ms<br>delta: 61.085s</span></td>
      </tr>
      <tr>
        <td>descendants_at_distance</td>
        <td class="red">0.007*<span class="tooltip">nx: 26.257us<br>nxcg: 3.593ms<br>delta: -3.567ms</span></td>
        <td class="red">0.003*<span class="tooltip">nx: 20.914us<br>nxcg: 6.145ms<br>delta: -6.124ms</span></td>
        <td class="red">0.001*<span class="tooltip">nx: 32.306us<br>nxcg: 30.265ms<br>delta: -30.233ms</span></td>
        <td class="red">0.065*<span class="tooltip">nx: 5.894ms<br>nxcg: 90.867ms<br>delta: -84.973ms</span></td>
      </tr>
      <tr>
        <td>edge_betweenness_centrality, k=10</td>
        <td class="red">0.167*<span class="tooltip">nx: 7.114ms<br>nxcg: 42.690ms<br>delta: -35.576ms</span></td>
        <td class="green">14.605<span class="tooltip">nx: 18.773s<br>nxcg: 1.285s<br>delta: 17.488s</span></td>
        <td class="green">10.153<span class="tooltip">nx: 121.627s<br>nxcg: 11.979s<br>delta: 109.647s</span></td>
        <td class="green">16.778<span class="tooltip">nx: 909.093s<br>nxcg: 54.182s<br>delta: 854.911s</span></td>
      </tr>
      <tr>
        <td>edge_betweenness_centrality, k=100</td>
        <td class="red">0.15*<span class="tooltip">nx: 43.901ms<br>nxcg: 292.226ms<br>delta: -248.325ms</span></td>
        <td class="green">24.172<span class="tooltip">nx: 172.498s<br>nxcg: 7.136s<br>delta: 165.361s</span></td>
        <td class="green">68.332<span class="tooltip">nx: 972.557s<br>nxcg: 14.233s<br>delta: 958.325s</span></td>
        <td class="green">112.81<span class="tooltip">nx: 8546.360s<br>nxcg: 75.759s<br>delta: 8470.601s</span></td>
      </tr>
      <tr>
        <td>ego_graph</td>
        <td class="red">0.726*<span class="tooltip">nx: 5.584ms<br>nxcg: 7.693ms<br>delta: -2.109ms</span></td>
        <td class="green">104.093<span class="tooltip">nx: 5.857s<br>nxcg: 56.265ms<br>delta: 5.801s</span></td>
        <td class="red">0.008*<span class="tooltip">nx: 1.431ms<br>nxcg: 187.666ms<br>delta: -186.235ms</span></td>
        <td class="green">150.036<span class="tooltip">nx: 302.519s<br>nxcg: 2.016s<br>delta: 300.503s</span></td>
      </tr>
      <tr>
        <td>eigenvector_centrality</td>
        <td class="green">7.409*<span class="tooltip">nx: 28.642ms<br>nxcg: 3.866ms<br>delta: 24.776ms</span></td>
        <td>--</td>
        <td class="green">263.997<span class="tooltip">nx: 242.879s<br>nxcg: 920.007ms<br>delta: 241.959s</span></td>
        <td class="green">376.161<span class="tooltip">nx: 458.153s<br>nxcg: 1.218s<br>delta: 456.935s</span></td>
      </tr>
      <tr>
        <td>generic_bfs_edges</td>
        <td class="red">0.032*<span class="tooltip">nx: 276.460us<br>nxcg: 8.570ms<br>delta: -8.294ms</span></td>
        <td class="red">0.38<span class="tooltip">nx: 1.238s<br>nxcg: 3.256s<br>delta: -2.018s</span></td>
        <td class="red">0.003*<span class="tooltip">nx: 109.543us<br>nxcg: 35.921ms<br>delta: -35.811ms</span></td>
        <td class="yellow">2.129<span class="tooltip">nx: 75.324s<br>nxcg: 35.374s<br>delta: 39.949s</span></td>
      </tr>
      <tr>
        <td>hits</td>
        <td class="yellow">1.703*<span class="tooltip">nx: 7.290ms<br>nxcg: 4.280ms<br>delta: 3.010ms</span></td>
        <td>--</td>
        <td class="green">33.932<span class="tooltip">nx: 64.038s<br>nxcg: 1.887s<br>delta: 62.150s</span></td>
        <td class="green">87.124<span class="tooltip">nx: 233.893s<br>nxcg: 2.685s<br>delta: 231.209s</span></td>
      </tr>
      <tr>
        <td>in_degree_centrality</td>
        <td class="red">0.459<span class="tooltip">nx: 293.520us<br>nxcg: 639.090us<br>delta: -345.570us</span></td>
        <td class="yellow">2.095*<span class="tooltip">nx: 57.231ms<br>nxcg: 27.314ms<br>delta: 29.917ms</span></td>
        <td class="yellow">1.645*<span class="tooltip">nx: 1.275s<br>nxcg: 775.082ms<br>delta: 499.672ms</span></td>
        <td class="yellow">1.808*<span class="tooltip">nx: 1.798s<br>nxcg: 994.536ms<br>delta: 803.470ms</span></td>
      </tr>
      <tr>
        <td>is_weakly_connected</td>
        <td class="red">0.001*<span class="tooltip">nx: 28.767us<br>nxcg: 31.511ms<br>delta: -31.482ms</span></td>
        <td class="green">9.442*<span class="tooltip">nx: 469.725ms<br>nxcg: 49.747ms<br>delta: 419.978ms</span></td>
        <td class="green">52.655<span class="tooltip">nx: 10.313s<br>nxcg: 195.850ms<br>delta: 10.117s</span></td>
        <td class="green">64.03<span class="tooltip">nx: 22.803s<br>nxcg: 356.123ms<br>delta: 22.446s</span></td>
      </tr>
      <tr>
        <td>k_truss</td>
        <td class="green">8.31*<span class="tooltip">nx: 12.342ms<br>nxcg: 1.485ms<br>delta: 10.857ms</span></td>
        <td class="green">539.652<span class="tooltip">nx: 6.103s<br>nxcg: 11.309ms<br>delta: 6.092s</span></td>
        <td>--</td>
        <td>--</td>
      </tr>
      <tr>
        <td>katz_centrality</td>
        <td class="green">122.153<span class="tooltip">nx: 4.219s<br>nxcg: 34.535ms<br>delta: 4.184s</span></td>
        <td class="green">648.055<span class="tooltip">nx: 24.870s<br>nxcg: 38.376ms<br>delta: 24.831s</span></td>
        <td class="green">300.147<span class="tooltip">nx: 274.585s<br>nxcg: 914.835ms<br>delta: 273.670s</span></td>
        <td class="green">24767.824<span class="tooltip">nx: 76304.239s<br>nxcg: 3.081s<br>delta: 76301.159s</span></td>
      </tr>
      <tr>
        <td>louvain_communities</td>
        <td class="yellow">2.479*<span class="tooltip">nx: 72.361ms<br>nxcg: 29.185ms<br>delta: 43.176ms</span></td>
        <td class="green">103.205<span class="tooltip">nx: 43.659s<br>nxcg: 423.034ms<br>delta: 43.236s</span></td>
        <td class="green">272.652<span class="tooltip">nx: 2409.300s<br>nxcg: 8.837s<br>delta: 2400.464s</span></td>
        <td class="green">208.843<span class="tooltip">nx: 4291.899s<br>nxcg: 20.551s<br>delta: 4271.348s</span></td>
      </tr>
      <tr>
        <td>number_weakly_connected_components</td>
        <td class="red">0.041*<span class="tooltip">nx: 1.296ms<br>nxcg: 31.745ms<br>delta: -30.449ms</span></td>
        <td class="green">9.956*<span class="tooltip">nx: 493.798ms<br>nxcg: 49.599ms<br>delta: 444.199ms</span></td>
        <td class="green">54.189<span class="tooltip">nx: 10.795s<br>nxcg: 199.216ms<br>delta: 10.596s</span></td>
        <td class="green">64.136<span class="tooltip">nx: 23.510s<br>nxcg: 366.558ms<br>delta: 23.143s</span></td>
      </tr>
      <tr>
        <td>out_degree_centrality</td>
        <td class="red">0.453<span class="tooltip">nx: 287.390us<br>nxcg: 634.436us<br>delta: -347.046us</span></td>
        <td class="yellow">2.156*<span class="tooltip">nx: 58.691ms<br>nxcg: 27.227ms<br>delta: 31.464ms</span></td>
        <td class="yellow">1.549*<span class="tooltip">nx: 1.202s<br>nxcg: 775.760ms<br>delta: 426.122ms</span></td>
        <td class="yellow">1.727*<span class="tooltip">nx: 1.714s<br>nxcg: 992.363ms<br>delta: 721.574ms</span></td>
      </tr>
      <tr>
        <td>overall_reciprocity</td>
        <td class="green">7.901*<span class="tooltip">nx: 15.359ms<br>nxcg: 1.944ms<br>delta: 13.415ms</span></td>
        <td class="green">563.312<span class="tooltip">nx: 4.525s<br>nxcg: 8.032ms<br>delta: 4.517s</span></td>
        <td class="green">1524.889<span class="tooltip">nx: 78.576s<br>nxcg: 51.529ms<br>delta: 78.524s</span></td>
        <td class="green">2257.021<span class="tooltip">nx: 239.907s<br>nxcg: 106.294ms<br>delta: 239.801s</span></td>
      </tr>
      <tr>
        <td>pagerank</td>
        <td class="yellow">1.696*<span class="tooltip">nx: 6.589ms<br>nxcg: 3.885ms<br>delta: 2.704ms</span></td>
        <td class="green">76.288<span class="tooltip">nx: 2.828s<br>nxcg: 37.065ms<br>delta: 2.791s</span></td>
        <td class="green">82.128<span class="tooltip">nx: 72.134s<br>nxcg: 878.314ms<br>delta: 71.256s</span></td>
        <td class="green">188.302<span class="tooltip">nx: 223.766s<br>nxcg: 1.188s<br>delta: 222.578s</span></td>
      </tr>
      <tr>
        <td>pagerank_personalized</td>
        <td class="yellow">1.406*<span class="tooltip">nx: 6.942ms<br>nxcg: 4.939ms<br>delta: 2.003ms</span></td>
        <td class="green">55.666<span class="tooltip">nx: 3.265s<br>nxcg: 58.648ms<br>delta: 3.206s</span></td>
        <td class="green">46.6<span class="tooltip">nx: 65.932s<br>nxcg: 1.415s<br>delta: 64.517s</span></td>
        <td class="green">120.002<span class="tooltip">nx: 237.002s<br>nxcg: 1.975s<br>delta: 235.027s</span></td>
      </tr>
      <tr>
        <td>reciprocity</td>
        <td class="red">0.01*<span class="tooltip">nx: 22.615us<br>nxcg: 2.313ms<br>delta: -2.291ms</span></td>
        <td class="red">0.013*<span class="tooltip">nx: 43.303us<br>nxcg: 3.431ms<br>delta: -3.388ms</span></td>
        <td class="red">0.009*<span class="tooltip">nx: 54.886us<br>nxcg: 5.864ms<br>delta: -5.809ms</span></td>
        <td class="red">0.172*<span class="tooltip">nx: 2.705ms<br>nxcg: 15.737ms<br>delta: -13.032ms</span></td>
      </tr>
      <tr>
        <td>shortest_path</td>
        <td class="red">0.064*<span class="tooltip">nx: 393.284us<br>nxcg: 6.105ms<br>delta: -5.711ms</span></td>
        <td class="yellow">1.1*<span class="tooltip">nx: 960.214ms<br>nxcg: 872.857ms<br>delta: 87.358ms</span></td>
        <td class="red">0.005*<span class="tooltip">nx: 168.884us<br>nxcg: 33.088ms<br>delta: -32.919ms</span></td>
        <td class="yellow">1.344<span class="tooltip">nx: 49.690s<br>nxcg: 36.964s<br>delta: 12.726s</span></td>
      </tr>
      <tr>
        <td>single_source_shortest_path_length</td>
        <td class="red">0.042*<span class="tooltip">nx: 235.942us<br>nxcg: 5.613ms<br>delta: -5.377ms</span></td>
        <td class="green">6.285*<span class="tooltip">nx: 280.148ms<br>nxcg: 44.572ms<br>delta: 235.576ms</span></td>
        <td class="red">0.003*<span class="tooltip">nx: 89.561us<br>nxcg: 31.926ms<br>delta: -31.836ms</span></td>
        <td class="green">14.241<span class="tooltip">nx: 15.347s<br>nxcg: 1.078s<br>delta: 14.269s</span></td>
      </tr>
      <tr>
        <td>single_target_shortest_path_length</td>
        <td class="red">0.042*<span class="tooltip">nx: 240.253us<br>nxcg: 5.682ms<br>delta: -5.442ms</span></td>
        <td class="green">5.482*<span class="tooltip">nx: 290.073ms<br>nxcg: 52.910ms<br>delta: 237.162ms</span></td>
        <td class="red">0.098*<span class="tooltip">nx: 3.321ms<br>nxcg: 33.767ms<br>delta: -30.446ms</span></td>
        <td class="green">7.672<span class="tooltip">nx: 11.358s<br>nxcg: 1.480s<br>delta: 9.878s</span></td>
      </tr>
      <tr>
        <td>transitivity</td>
        <td class="yellow">1.682*<span class="tooltip">nx: 15.338ms<br>nxcg: 9.116ms<br>delta: 6.222ms</span></td>
        <td class="green">331.111<span class="tooltip">nx: 7.525s<br>nxcg: 22.726ms<br>delta: 7.502s</span></td>
        <td class="green">937.121<span class="tooltip">nx: 196.579s<br>nxcg: 209.769ms<br>delta: 196.369s</span></td>
        <td class="green">2831.727<span class="tooltip">nx: 2157.310s<br>nxcg: 761.836ms<br>delta: 2156.548s</span></td>
      </tr>
      <tr>
        <td>triangles</td>
        <td class="red">0.625*<span class="tooltip">nx: 5.069ms<br>nxcg: 8.105ms<br>delta: -3.036ms</span></td>
        <td class="green">56.354<span class="tooltip">nx: 2.591s<br>nxcg: 45.977ms<br>delta: 2.545s</span></td>
        <td class="green">55.831<span class="tooltip">nx: 55.304s<br>nxcg: 990.560ms<br>delta: 54.313s</span></td>
        <td class="green">124.634<span class="tooltip">nx: 220.697s<br>nxcg: 1.771s<br>delta: 218.926s</span></td>
      </tr>
      <tr>
        <td>weakly_connected_components</td>
        <td class="red">0.035*<span class="tooltip">nx: 1.275ms<br>nxcg: 36.601ms<br>delta: -35.326ms</span></td>
        <td class="green">7.677*<span class="tooltip">nx: 493.186ms<br>nxcg: 64.240ms<br>delta: 428.946ms</span></td>
        <td class="green">11.559<span class="tooltip">nx: 10.928s<br>nxcg: 945.427ms<br>delta: 9.983s</span></td>
        <td class="green">20.597<span class="tooltip">nx: 24.027s<br>nxcg: 1.167s<br>delta: 22.861s</span></td>
      </tr>
    </tbody>
  </table>
  <br>
  <p style="font-size: 12px; text-indent: 20px;">
    * indicates that the time difference between CPU and GPU was insignificant (<1s)
  </p>

--------------

.. raw:: html

   <style>
   .meta-footer {
       font-size: 12px;
       line-height: 1.1;
       color: #666;
   }
   </style>

.. container:: meta-footer

   Benchmarks were run on a system with the following specifications:

   - **NetworkX Version**: 3.4.2
   - **nx-cugraph Version**: 25.02.00a36
   - **CPU**: Intel(R) Xeon(R) w9-3495X (56 cores)
   - **GPU**: 1 x NVIDIA GeForce RTX 3090 (24.0 GB)
   - **CUDA Version**: 12.8
   - **RAM**: 251.18 GB
   - **OS**: #115-Ubuntu SMP Mon Apr 15 09:52:04 UTC 2024