AmatyaAvadhanula commented on PR #12419: URL: https://github.com/apache/druid/pull/12419#issuecomment-1101994710
> > Could you also please share any insights to estimate a ballpark performance change? > > If this work is inspired by a real world scenario, you could build a performance test or benchmark similar to that scenario. IMO that's the best way to approach this kind of thing, because otherwise it can be tough to make sure you're attacking an actual bottleneck. The cost in a "stable" state will reduce significantly. The bottleneck is startup with a large number of segments or when there are several segments waiting to be loaded / dropped. I do have an approach in mind and will try to run a benchmark simulating these scenarios -- 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: [email protected] For queries about this service, please contact Infrastructure at: [email protected] --------------------------------------------------------------------- To unsubscribe, e-mail: [email protected] For additional commands, e-mail: [email protected]
