Github user mateiz commented on the pull request:
https://github.com/apache/spark/pull/165#issuecomment-37902896
Both look like good fixes -- I didn't realize before that each RDD would be
used twice. This will actually make us very competitive with GraphLab for
example in benchmarks.
One thing we may want to do later is to make the storage level for these
persists configurable, but it doesn't seem necessary for now. In general
managing memory use by libraries will be an interesting question.
---
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.
---
