Github user vrilleup commented on the pull request:
https://github.com/apache/spark/pull/964#issuecomment-48001763
@mengxr I tested a matrix today where n = 300k, and the aggregate took much
longer time. I think it becomes master node network bound in this case. 300k *
500 (#partitions) doubles need to be sent to the master node. Could you point
me the tree aggregate that you mentioned? Thank you!
---
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.
---
