You are viewing a plain text version of this content. The canonical link for it is here.
Posted to commits@beam.apache.org by me...@apache.org on 2018/07/23 07:29:50 UTC
[beam-site] 01/03: [BEAM-4827] Add Nexmark dashboards to nexmark
page on the website
This is an automated email from the ASF dual-hosted git repository.
mergebot-role pushed a commit to branch mergebot
in repository https://gitbox.apache.org/repos/asf/beam-site.git
commit fabc1e331784f2c550a8ac252eabca34ab60cf9a
Author: Etienne Chauchot <ec...@apache.org>
AuthorDate: Thu Jul 19 14:41:08 2018 +0200
[BEAM-4827] Add Nexmark dashboards to nexmark page on the website
---
src/documentation/sdks/nexmark.md | 45 +++++++++++++++++++++++++++++++++++++++
1 file changed, 45 insertions(+)
diff --git a/src/documentation/sdks/nexmark.md b/src/documentation/sdks/nexmark.md
index 8e9c6a7..be96a58 100644
--- a/src/documentation/sdks/nexmark.md
+++ b/src/documentation/sdks/nexmark.md
@@ -653,3 +653,48 @@ Submit to the cluster:
--streaming=false \
--manageResources=false \
--monitorJobs=true"
+
+## Nexmark dashboards
+Bellow dashboards are used as a CI mechanism to ensure no-regression on the Beam components. They are not supposed to be benchmark comparision of the runners or engines. Especially because:
+- Parameters of the runners are not the same
+- Nexmark is run with the runners in local (most of the time embedded) mode
+- Nexmark run on a shared machine that also run all the CI and build.
+- Runners have different support of the Beam model
+- Runners have different strengths that make comparision difficult:
+ - Some runners were designed to be batch oriented, others streaming oriented
+ - Some are designed towards sub-second latency, others support auto-scaling
+
+### Dashboards content
+At each commit on master, Nexmark suites are run and plots are created on the graphs.
+
+There are 2 kinds of dashboards:
+- one for performances (run times of the queries)
+- one for the size of the output PCollection (which should be constant)
+
+There are dashboards for these runners (others to come):
+- spark
+- flink
+- direct runner
+
+Each dashboard contains:
+- graphs in batch mode
+- graphs in streaming mode
+- graphs for the 13 queries.
+
+### Performance dashboards links
+
+[Nexmark performance direct runner](https://apache-beam-testing.appspot.com/explore?dashboard=5084698770407424)
+
+[Nexmark performance flink runner](https://apache-beam-testing.appspot.com/explore?dashboard=5699257587728384)
+
+[Nexmark performance spark runner](https://apache-beam-testing.appspot.com/explore?dashboard=5138380291571712)
+
+
+### Output size dashboards links
+
+[Nexmark output size direct runner](https://apache-beam-testing.appspot.com/explore?dashboard=5099379773931520)
+
+[Nexmark output size flink runner](https://apache-beam-testing.appspot.com/explore?dashboard=5731568492478464)
+
+[Nexmark output size spark runner](https://apache-beam-testing.appspot.com/explore?dashboard=5163657986048000)
+