sgup432 commented on PR #15803: URL: https://github.com/apache/lucene/pull/15803#issuecomment-4027716459
>Are there any existing call points in Lucene that might benefit from this? Maybe DirectoryTaxonomyReader#getBulkPath? Maybe create a spin-off issue to move to using this where it makes sense, or make that change as part of this? I think we should also make the change to integrate the solution as part of this? As otherwise it looks like a solution looking for a problem. Also I am thinking if we can add a benchmark(JMH also fine) around this(or run a existing one), and showcase where it shines for desired N(hits) and number of segments. Some worst cases I can think for this: very few hits and many segments, or all hits in 1 segment? Also does this change benefit when we have concurrent search enabled? -- 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]
