This is an automated email from the ASF dual-hosted git repository. mergebot-role pushed a commit to branch asf-site in repository https://gitbox.apache.org/repos/asf/beam-site.git
commit 671e21752bd93ffba5dd2d86f08d64697e26cc30 Author: Mergebot <[email protected]> AuthorDate: Mon Jul 23 07:32:25 2018 +0000 Prepare repository for deployment. --- content/documentation/sdks/java/nexmark/index.html | 62 ++++++++++++++++++++++ 1 file changed, 62 insertions(+) diff --git a/content/documentation/sdks/java/nexmark/index.html b/content/documentation/sdks/java/nexmark/index.html index dc3a5b5..96349f1 100644 --- a/content/documentation/sdks/java/nexmark/index.html +++ b/content/documentation/sdks/java/nexmark/index.html @@ -263,6 +263,13 @@ <li><a href="#running-query-0-on-a-spark-cluster-with-apache-hadoop-yarn">Running query 0 on a Spark cluster with Apache Hadoop YARN</a></li> </ul> </li> + <li><a href="#nexmark-dashboards">Nexmark dashboards</a> + <ul> + <li><a href="#dashboards-content">Dashboards content</a></li> + <li><a href="#performance-dashboards-links">Performance dashboards links</a></li> + <li><a href="#output-size-dashboards-links">Output size dashboards links</a></li> + </ul> + </li> </ul> @@ -989,6 +996,61 @@ PUBSUB_TOPCI=<existing pubsub topic> </code></pre> </div> +<h2 id="nexmark-dashboards">Nexmark dashboards</h2> +<p>Below dashboards are used as a CI mechanism to detect no-regression on the Beam components. They are not supposed to be benchmark comparision of the runners or engines. Especially because:</p> +<ul> + <li>Parameters of the runners are not the same</li> + <li>Nexmark is run with the runners in local (most of the time embedded) mode</li> + <li>Nexmark runs on a shared machine that also run all the CI and build.</li> + <li>Runners have different support of the Beam model</li> + <li>Runners have different strengths that make comparison difficult: + <ul> + <li>Some runners were designed to be batch oriented, others streaming oriented</li> + <li>Some are designed towards sub-second latency, others support auto-scaling</li> + </ul> + </li> +</ul> + +<h3 id="dashboards-content">Dashboards content</h3> +<p>At each commit on master, Nexmark suites are run and plots are created on the graphs.</p> + +<p>There are 2 kinds of dashboards:</p> +<ul> + <li>one for performances (run times of the queries)</li> + <li>one for the size of the output PCollection (which should be constant)</li> +</ul> + +<p>There are dashboards for these runners (others to come):</p> +<ul> + <li>spark</li> + <li>flink</li> + <li>direct runner</li> +</ul> + +<p>Each dashboard contains:</p> +<ul> + <li>graphs in batch mode</li> + <li>graphs in streaming mode</li> + <li>graphs for the 13 queries.</li> +</ul> + +<h3 id="performance-dashboards-links">Performance dashboards links</h3> + +<p><a href="https://apache-beam-testing.appspot.com/explore?dashboard=5084698770407424";>Nexmark performance direct runner</a></p> + +<p><a href="https://apache-beam-testing.appspot.com/explore?dashboard=5699257587728384";>Nexmark performance flink runner</a></p> + +<p><a href="https://apache-beam-testing.appspot.com/explore?dashboard=5138380291571712";>Nexmark performance spark runner</a></p> + +<h3 id="output-size-dashboards-links">Output size dashboards links</h3> + +<p><a href="https://apache-beam-testing.appspot.com/explore?dashboard=5099379773931520";>Nexmark output size direct runner</a></p> + +<p><a href="https://apache-beam-testing.appspot.com/explore?dashboard=5731568492478464";>Nexmark output size flink runner</a></p> + +<p><a href="https://apache-beam-testing.appspot.com/explore?dashboard=5163657986048000";>Nexmark output size spark runner</a></p> + + </div> </div> <!--
