Github user CodingCat commented on the pull request:
https://github.com/apache/spark/pull/1313#issuecomment-50292050
Yes, I understand that
but in that case the behaviour is exactly the same as the code in the
commit
(https://github.com/CodingCat/spark/commit/f9900141d10168d0e09f4daef85a6459fe40a254#diff-bad3987c83bd22d46416d3dd9d208e76R350),
which were submitted several days ago. There, I just moved the noPref branch
inside if (NODE_LOCAL is allowed),
In last round of discussion, @kayousterhout @mateiz @mridulm @lirui-intel
pointed out that: that version of PR cannot resolve the situation that if we
have process_local + nopref tasks and we have launched all PROCESS_LOCAL
tasks, we still need to pass through PROCESS_LOCAL, (3s), NODE_LOCAL + NOPREF,
where the 3s is unnecessary
The current PR can avoid this 3s, the remaining issue might be that, when
we call with NOPREF, it may return a NODE_LOCAL task
---
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.
---
