jiaan.geng created SPARK-44781:
----------------------------------
Summary: Runtime filter should supports reuse exchange if it can
reduce the data size of application side
Key: SPARK-44781
URL: https://issues.apache.org/jira/browse/SPARK-44781
Project: Spark
Issue Type: New Feature
Components: SQL
Affects Versions: 4.0.0
Reporter: jiaan.geng
Currently, Spark runtime filter only supports using the subquery on one table.
In fact, we can reuse the exchange, even if it is a shuffle exchange.
If the shuffle exchange come from a join which has one side with selective
predicates, so the results of the join can be used to prune the data amount of
the application side.
--
This message was sent by Atlassian Jira
(v8.20.10#820010)
---------------------------------------------------------------------
To unsubscribe, e-mail: [email protected]
For additional commands, e-mail: [email protected]
