Github user srowen commented on the pull request:
https://github.com/apache/spark/pull/4234#issuecomment-71809831
@ankurdave Ah of course. Hm, the thing is there is a loop of
transformations here, so in theory you'd have to save up all the RDD references
and unpersist at the very end -- but even that may be insufficient as the
return value isn't even materialized before it's returned. Also, operations
like `outerJoinVertices` seem to `cache()` vertices internally too. It's a bit
like the old days of memory management!
What's the right thing here? don't `cache()` in this method at all since
lower layers do some of that management? force materialization?
---
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]
