viirya commented on pull request #32136:
URL: https://github.com/apache/spark/pull/32136#issuecomment-826562394
Hmm, does it mean this is an invalid requirement for Spark and we may not
address it? I hope not the case.
I totally agree that there are many aspects needed to be carefully looked
at. We should look at these aspects specially.
Let me figure out. Is this a real issue? Yes. As our production applications
face checkpoint related issues, the capacity to make state store location
stable is one step towards fixing the issues. Should we address the issues? I
think so, except that we don't expect to improve Spark used in streaming
use-cases. Is this interface good enough? Maybe not, I'm open to hear
suggestions about how/what we can make it better/useful to us and others.
--
This is an automated message from the Apache Git Service.
To respond to the message, please log on to GitHub and use the
URL above to go to the specific comment.
For queries about this service, please contact Infrastructure at:
[email protected]
---------------------------------------------------------------------
To unsubscribe, e-mail: [email protected]
For additional commands, e-mail: [email protected]
