Github user jerryshao commented on the pull request:
https://github.com/apache/spark/pull/8976#issuecomment-146384050
With [SPARK-4352](https://issues.apache.org/jira/browse/SPARK-4352) getting
merged, this `SparkContext.preferredNodeLocationData` can easily be achieved by
computing the locality information at the start of AM and issue
`ContainerRequest` with locality information.
My concern is that for some workloads like "terasort", "wordcount",
normally we will not enable dynamic allocation to do the benchmark, so there's
no way to specify the preferred locality and the test result may not be so good.
---
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]
