jpountz opened a new issue, #12439: URL: https://github.com/apache/lucene/issues/12439
### Description With Elasticsearch, we have been observing that queries with many SHOULD clauses can get multiple times faster with our brute-force BS1 scorer compared to our scorers that support dynamic pruning and get used by default. I don't think we can reliably predict when BS1 would perform better, but we could run luceneutil with increasing numbers of clauses (2, 4, 8, 16, 32) of various term frequencies and see when BS1 performs better. This would give us a threshold that would likely be imperfect but would also be better than nothing? -- 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.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
