Github user HeartSaVioR commented on the issue:
https://github.com/apache/storm/pull/2270
@Ethanlm
Thanks for doing following up test. Looks good to me.
Looks like we can put more effort for further optimizations, but I think it
could be handled as follow up issue, especially if you are working on follow-up
tasks. We're dealing with network latency which is much bigger than what
optimization can bring to.
Btw, when you are doing the test with 1 worker, you don't need to test both
LocalOrShuffle and Shuffle, since LocalOrShuffle is just a shuffle with
limiting tasks to local if possible. If you still would want to play with
performance test, you may also want to play with #2261 given that it will just
replace the number for LoadAwareShuffleGrouping.
---
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.
---
