mikemccand commented on PR #13472: URL: https://github.com/apache/lucene/pull/13472#issuecomment-2185015445
> So searching concurrently results in a lot more duplicate work. Thanks @jpountz -- I had actually not realized this. I thought concurrent search exchanged information across each thread/slice and was able to be more efficient (visit/collect fewer hits, spend less aggregated CPU) than sequential search. Conceptually it seems like Lucene should eventually be able to be more efficient when collecting concurrently. > This was one of my motivations for enabling concurrency on nightly benchmarks: so that we identify these cases and fix them. +1, I think Lucene has a lot of exciting innovations still in this area! -- 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
