Github user revans2 commented on the pull request:
https://github.com/apache/storm/pull/765#issuecomment-150247138
@mjsax If you want to take some time to debug things for a better
comparison I am happy to wait. It is entirely possible that I somehow messed
things up when I was enabling hybrid-mode so getting and apples to apples
comparison is important. I want the best final solution no matter what it is,
but I also don't want to wait forever. For efficiency and cluster utilization
purposes I really would like to pull in batching into the clusters that I run,
but because the configuration/interfaces are incompatible with one another
between the two approaches I don't want to do it until open source has decided
on a direction they want to go in.
In the mean time I am going to be working on improving the metrics
gathering. You said previously that with your first batching attempt you saw
up to 6x the throughput. I suspect a lot of that is because the metrics are so
very very slow.
---
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.
---
