alamb commented on issue #17171: URL: https://github.com/apache/datafusion/issues/17171#issuecomment-3267280510
> My counter argument to this would be that this is only a problem if the size of your build side ≈ the size of your probe side, but if that's the case you already probably have a suboptimal or slow query (and maybe a hash join wasn't even the right choice). If your build side is small then the overhead of building a bloom filter should be small relative to the speedup you get if lookups are even 15% faster than a hash table. Maybe we can dynamically choose here too -- like if the bloom filter size isn't big enough for the number of distinct values then stop building it or 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: github-unsubscr...@datafusion.apache.org For queries about this service, please contact Infrastructure at: us...@infra.apache.org --------------------------------------------------------------------- To unsubscribe, e-mail: github-unsubscr...@datafusion.apache.org For additional commands, e-mail: github-h...@datafusion.apache.org
