Github user staple commented on the pull request: https://github.com/apache/spark/pull/1362#issuecomment-49457211 I added some patches to address the above comments and introduce a timed test. My test uses an RDD with no preferred locations in the entire dependency graph. The reason for this is that some RDD types employ mitigations for the exponential blowup problem by overriding getPreferredLocations to return the preferred locations of their parents. The easiest way to circumvent these mitigations and implement a failing test was to create a dependency graph with no preferred locations.
--- 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. ---