manuzhang edited a comment on issue #26813: [SPARK-30188][SQL][WIP] Enable adaptive query execution by default URL: https://github.com/apache/spark/pull/26813#issuecomment-564350695 @tgravescs Yes, we've been back-porting AQE patches to 2.3 branch and running nightly tpcds tests. I've seen a regression with [tpcds q87](https://github.com/databricks/spark-sql-perf/blob/master/src/main/resources/tpcds_2_4/q87.sql) on 1TB data with AQE enabled. The regression was due to executor GC issue brought by `ReduceNumShufflePartitions` and setting `spark.sql.adaptive.shuffle.targetPostShuffleInputSize=16777216` could alleviate it.
---------------------------------------------------------------- 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. For queries about this service, please contact Infrastructure at: [email protected] With regards, Apache Git Services --------------------------------------------------------------------- To unsubscribe, e-mail: [email protected] For additional commands, e-mail: [email protected]
