Github user mateiz commented on the pull request: https://github.com/apache/spark/pull/694#issuecomment-43038729 BTW I kind of prefer this solution over the alternative one, at least for now. The very best solution would be to broadcast the part of the task description that's shared across tasks to all nodes. Otherwise putting these in the block store will lead to lots of replicated data. But it's better to warn for now and maybe tell them to use broadcast.
--- 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. ---