Github user sryza commented on the pull request:
https://github.com/apache/spark/pull/634#issuecomment-49126877
That's a good point. Changing it for YARN seems like the right thing, and
80% sounds reasonable to me.
Another thing is the wait time. Previously it was 6 seconds, but now
spark.scheduler.maxRegisteredExecutorsWaitingTime defaults to 5 times that. 30
seconds seems a little excessive to me in general - at least for jobs without
caching, after a couple seconds the wait outweighs scheduling some non-local
tasks. What do you think about decreasing this to 6 seconds in general? Or at
least for YARN.
---
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 infrastruct...@apache.org or file a JIRA ticket
with INFRA.
---
