Github user pwendell commented on the pull request:
https://github.com/apache/spark/pull/3026#issuecomment-61222814
Hey @tdas I did a first pass on this. This makes sense to me overall and
seems like a straightforward next step in the design. I had a few comments
about naming. The main issue I noticed is that having a single function
`getOrAllocateBlocksToBatch` was very hard to reason about and the semantics
seemed like they were lost up the call chain. I wonder if it can be improved.
All that said, I was just reviewing this slice of a larger change and I
haven't been able yet to sit down and I haven't done a thorough vetting of the
larger fault tolerance guarentees (e.g. think about corner cases during
failures, etc). I will try to do that tomorrow.
---
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]
