uschindler commented on issue #12358: URL: https://github.com/apache/lucene/issues/12358#issuecomment-1585286995
I checked the repository. When Tantitvy says it is 2 times faster then this is also unfair: It does not warm up the JVM. It runs the list of queries in cold, unoptimized state - COLD! Very unfair, sorry. See code: - https://github.com/Tony-X/search-benchmark-game/blob/a6f71237e84aeb47ee8f5b923f16b760b032e1fc/engines/lucene-9.5.0/src/main/java/DoQuery.java - https://github.com/Tony-X/search-benchmark-game/blob/a6f71237e84aeb47ee8f5b923f16b760b032e1fc/engines/lucene-9.5.0/query.sh (no extra options passed) So basically it runs the queries on a cold JVM so the first queries in the list are all running in interpreted mode, especially when the new vector features or mmapdir are used. Did I oversee something? -- 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
