Github user jaceklaskowski commented on the issue:
https://github.com/apache/spark/pull/16250
Just to have the list of the reasons not to accept the changes:
1. @rxin "just fyi this is going to be slower than the original code."
2. @shivaram "Its hard to profile these things closely enough to figure out
how much overhead it adds"
3. @srowen "There is a tiny risk of breaking something or regressing
performance."
4. @shivaram "I don't think the improvement in readability is worth the
risk in terms of performance regression"
I however think the points above are the exact reasons why you could accept
this and the following changes to DAGScheduler. We should *not* be afraid of
making changes to the core. We've got no evidence the points above hold and
that's what worries me the most.
Thanks for your help anyway! It was worth to have sent the PR to learn from
you!
---
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.
---
---------------------------------------------------------------------
To unsubscribe, e-mail: [email protected]
For additional commands, e-mail: [email protected]
