Github user davies commented on the pull request:
https://github.com/apache/spark/pull/11248#issuecomment-187448749
@nongli We already avoid the copy until found the match. In order to
support left N rows match with right M rows, we have to copy some of them.
Right now, we copy the keys from left, and the rows from right. It's possible
to have less copies, I'd like to leave that out of this PR, because SMJ usually
required Exchange and Sort, those are more expensive than coping the matched
rows from right.
---
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 [email protected] or file a JIRA ticket
with INFRA.
---
---------------------------------------------------------------------
To unsubscribe, e-mail: [email protected]
For additional commands, e-mail: [email protected]
