Github user yhuai commented on the issue: https://github.com/apache/spark/pull/15575 Our planner decides if to add an `ShuffleExchange` by consider `outputPartitioning` and `requiredDistribution` together. If the `outputPartitioning` of the child does not satisfy the `requiredDistribution` of the parent, we will add a `ShuffleExchange` operator. When we say `child`, it is literally the child of a node. It does not matter if the child is a `ShuffleExchange` or not. For a node, when its `outputPartitioning` is `child. outputPartitioning`, this node does not shuffle data.
--- If your project is set up for it, you can reply to this email and have your reply appear on GitHub as well. If your project does not have this feature enabled and wishes so, or if the feature is enabled but not working, please contact infrastructure at infrastruct...@apache.org or file a JIRA ticket with INFRA. --- --------------------------------------------------------------------- To unsubscribe, e-mail: reviews-unsubscr...@spark.apache.org For additional commands, e-mail: reviews-h...@spark.apache.org