zhaih opened a new pull request, #12660: URL: https://github.com/apache/lucene/pull/12660
### Description This PR is still a draft contains all kinds of magic number and names. It contains changes of #12651 so I would like to merge that one first. But I do think the logic is correct given the benchmark result below, and it seems promising as well. ### Test #### Set up 50MB writer buffer, force-merge to 1 at last #### Candidate ``` recall latency nDoc fanout maxConn beamWidth visited index ms Force merge done, time: 2 ms 0.838 0.28 10000 0 64 250 100 3649 1.00 post-filter Force merge done, time: 20012 ms 0.744 1.15 100000 0 64 250 100 66680 1.00 post-filter Force merge done, time: 470161 ms 0.686 12.94 1000000 0 64 250 100 1019300 1.00 post-filter ``` #### Baseline ``` recall latency nDoc fanout maxConn beamWidth visited index ms Force merge done, time: 3 ms 0.838 0.28 10000 0 64 250 100 3749 1.00 post-filter Force merge done, time: 97579 ms 0.745 1.19 100000 0 64 250 100 145355 1.00 post-filter Force merge done, time: 2059821 ms 0.683 12.81 1000000 0 64 250 100 2902514 1.00 post-filter ``` -- 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