Github user pwoody commented on the pull request:
https://github.com/apache/spark/pull/13178#issuecomment-220170469
I could be wrong, but I believe the main difference between this patch and
limiting the threadpool is that the broadcasts will still be retained in the
memory manager by the reference in the node (which should stick around until
execution is completed). The change here should get GC'd and properly evicted
from the memory manager as each node gets executed.
Re: bugfix v.s. improvement - this issue prevents certain large queries
(100s of broadcasts) from reasonably being completed with DataFrames. This
change might be an improvement, but I think there should be a backport of
something to 1.6.x to mitigate this issue.
---
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.
---
---------------------------------------------------------------------
To unsubscribe, e-mail: reviews-unsubscr...@spark.apache.org
For additional commands, e-mail: reviews-h...@spark.apache.org
