Github user andrewor14 commented on the pull request:
https://github.com/apache/spark/pull/7532#issuecomment-126878461
> I'm not sure I completely understand all the code here, but it seems to
me like this case would be covered by just ignoring the requested number of
cores in the "one executor per worker" case, and requesting as many cores as
the worker has available.
We can't ignore the requested number of cores because we need to support
`spark.cores.max` in the spread out case. E.g. we have 5 workers with 10 cores
each, if `spark.cores.max = 18` then we should end up with `[6, 6, 6, 5, 5]`.
We can't just allocate as all the available cores on the worker, so instead we
need to schedule the cores 1 by 1.
---
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]
