codaitya commented on pull request #446: URL: https://github.com/apache/lucene/pull/446#issuecomment-1054179398
Sorry for the delay in getting back to this, got busy with work and also needed time to study more details on how Lucene does segment merges. > Why do we need to exclude small segments from regular merges? The idea was that since writer threads can flush on their own, the new segments are eligible for regular merge. Regular merges can pick up these small segments, spend lot of time on these merges, and they might become unavailable for fullFlush merges. But I think this step (writer threads flush on their own) should normally only kick in once the RAMBuffer fills up, which would mean that the resulting segment isn’t that small. So I agree small segments need not be excluded from regular merges. I will update the PR to not override the `findMerges` function and just do the computation of small segment merge in `findFullFlushMerges` -- 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
