Github user koeninger commented on the pull request:
https://github.com/apache/spark/pull/3798#issuecomment-69446001
I went ahead and implemented locality and checkpointing of generated rdds.
Couple of points
- still depends on SPARK-4014 eventually being merged, for efficiency's
sake.
- I ran into classloader / class not found issues trying to checkpoint
KafkaRDDPartition directly. Current solution is to transform them to/from
tuples, ugly but it works. If you know what the issue is there, let me
know.
- I've got a use case that requires overriding the compute method on the
DStream (basically, modifying offsets to a fixed delay rather than "now").
I'm assuming you'd prefer a user supplied function to do the transformation
rather than subclassing, but let me know.
On Mon, Jan 5, 2015 at 7:59 PM, Tathagata Das <[email protected]>
wrote:
> Great! Keep me posted.
>
> â
> Reply to this email directly or view it on GitHub
> <https://github.com/apache/spark/pull/3798#issuecomment-68815205>.
>
---
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]
