You are viewing a plain text version of this content. The canonical link for it is here.
Posted to issues@lucene.apache.org by "neoremind (via GitHub)" <gi...@apache.org> on 2026/04/18 15:51:30 UTC

[I] Generalize jmh-table.py into a reusable visualization for any JMH benchmark [lucene]

neoremind opened a new issue, #15968:
URL: https://github.com/apache/lucene/issues/15968

   ### Description
   
   Spinoff from #15950 as @mikemccand suggested. The `jmh-table.py` visualization introduced in this PR for ScoreDoc sort benchmark is really cool, it's already a near-general solution. 
   
   One enhancement is: the script currently hardcodes `size` and `distribution` which ties specifically to `ScoreDocSortBenchmark`. It would be better to generalize this into a reusable tool that works with any of lucene's JMH benchmarks by auto-discovering `@Param` names.
   
   In addition, one new feature: JMH benchmarks produce wall-of-text results that can be hard to look by developers at a glance. It would be nicer to present them in a more visual layout using the same GUI that `jmh-table.py` generates, the layout looks similar to what @gsmiller demonstrated against `PartitionByLeafBenchmark` in #15938. 
   


-- 
This is an automated message from the Apache Git Service.
To respond to the message, please log on to GitHub and use the
URL above to go to the specific comment.

To unsubscribe, e-mail: issues-unsubscribe@lucene.apache.org.apache.org

For queries about this service, please contact Infrastructure at:
users@infra.apache.org


---------------------------------------------------------------------
To unsubscribe, e-mail: issues-unsubscribe@lucene.apache.org
For additional commands, e-mail: issues-help@lucene.apache.org


Re: [I] Generalize jmh-table.py into a reusable visualization for any JMH benchmark [lucene]

Posted by "mikemccand (via GitHub)" <gi...@apache.org>.
mikemccand commented on issue #15968:
URL: https://github.com/apache/lucene/issues/15968#issuecomment-4406499335

   This would be amazing, and it sounds like it's not so hard.  Then we'd have highly visual summary [like this](https://githubsearch.mikemccandless.com/jmh-sort-scoredoc-benchy.html#;dist=random)... visualizations are powerful because they leverage the amazing highly concurrent computer that is our visual system ... at a glance we can learn so much more, than if we stare and stare and stare at a big table of numbers.  It's kind of a form of mechanical sympathy....:
   
   <img width="1243" height="1154" alt="Image" src="https://github.com/user-attachments/assets/aeaeb204-387f-43ff-8a38-fbae7608de0e" />
   
   


-- 
This is an automated message from the Apache Git Service.
To respond to the message, please log on to GitHub and use the
URL above to go to the specific comment.

To unsubscribe, e-mail: issues-unsubscribe@lucene.apache.org

For queries about this service, please contact Infrastructure at:
users@infra.apache.org


---------------------------------------------------------------------
To unsubscribe, e-mail: issues-unsubscribe@lucene.apache.org
For additional commands, e-mail: issues-help@lucene.apache.org