Github user mengxr commented on the pull request:
https://github.com/apache/spark/pull/593#issuecomment-41932527
@tmyklebu I think the inconsistency in partitioners created more stages in
the computation and larger shuffle size. We should subclass `Partitioner` and
override `equals`, so Spark knows two RDDs use the same partitioner and then
group/join operations can take advantage of it. Hope it can help.
---
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.
---
