original-brownbear commented on PR #13864: URL: https://github.com/apache/lucene/pull/13864#issuecomment-2403376299
> I suspect that the main offender are doc values terms dictionaries Jup that's it :) > and the storage overhead would likely be negligible without the heap overhead of The thing is, we use all the arrays in `Meta` on the hot path when reading from `DirectMonotonic` reader, so just moving this thing off-heap means either taking a (potentially considerable looking at the profiling for this thing) performance hit or a rather complicated solution where we have off-heap meta and lazily set up the concrete arrays? Or are you suggesting only doing this for offsets? -- 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-unsubscr...@lucene.apache.org For queries about this service, please contact Infrastructure at: us...@infra.apache.org --------------------------------------------------------------------- To unsubscribe, e-mail: issues-unsubscr...@lucene.apache.org For additional commands, e-mail: issues-h...@lucene.apache.org
