Github user sryza commented on the pull request:
https://github.com/apache/spark/pull/894#issuecomment-44666824
Workers currently get 384 added as well (see
YarnAllocationHandler.MEMORY_OVERHEARD). At a first approximation, this seems
reasonable to me. The AM and executors are both JVMs loading most of the same
classes, so probably have similar fixed overheads. If we're adding configs, it
probably makes sense to have a separate one for each though.
---
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.
---
