You are viewing a plain text version of this content. The canonical link for it is here.
Posted to reviews@spark.apache.org by "cloud-fan (via GitHub)" <gi...@apache.org> on 2023/03/14 02:21:14 UTC

[GitHub] [spark] cloud-fan commented on pull request #40385: [SPARK-42753] ReusedExchange refers to non-existent nodes

cloud-fan commented on PR #40385:
URL: https://github.com/apache/spark/pull/40385#issuecomment-1467245909

   Yea AQE may remove materialized query stages due to optimizations like empty relation propagation, but I think it's fine as the shuffle files are still there (we don't unregister the shuffle), so the reused shuffle operator can still read these shuffle files using the shuffle id. The problem with EXPLAIN is it only looks for the referenced exchange in the query plan tree, I think we can also look up from the AQE stage cache map?


-- 
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: reviews-unsubscribe@spark.apache.org

For queries about this service, please contact Infrastructure at:
users@infra.apache.org


---------------------------------------------------------------------
To unsubscribe, e-mail: reviews-unsubscribe@spark.apache.org
For additional commands, e-mail: reviews-help@spark.apache.org