Github user jiangxb1987 commented on the issue: https://github.com/apache/spark/pull/20414 @felixcheung You are right that I didn't make it clear there should be still many shuffle blocks, and if you have the read task retried it should be slower than using `repartition(1)` directly. Now I tend to fix the issue following the latter fix-shuffle-fetch-order way, since it may resolve for general cases.
--- --------------------------------------------------------------------- To unsubscribe, e-mail: reviews-unsubscr...@spark.apache.org For additional commands, e-mail: reviews-h...@spark.apache.org