Github user dongjoon-hyun commented on the issue: https://github.com/apache/spark/pull/22348 Thank you for experiment, @SongYadong . IMO, if the gain is not obvious, we had better keep the current proven one because new one can introduce some future regressions (or even bugs).
--- --------------------------------------------------------------------- To unsubscribe, e-mail: reviews-unsubscr...@spark.apache.org For additional commands, e-mail: reviews-h...@spark.apache.org