Github user andrewor14 commented on the pull request:
https://github.com/apache/spark/pull/7770#issuecomment-127164852
To add to @pwendell's comment, the biggest advantage of accumulators is the
ability to deal with failed and resubmitted tasks. Double counting has been a
perennial issue for consuming data encoded in task metrics through the listener
interface. For this reason I think it's good to eventually migrate everything
to accumulators, which is also more general and has little impact on backwards
compatibility.
---
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]
