Github user arunmahadevan commented on the pull request:
https://github.com/apache/storm/pull/1217#issuecomment-200669173
@roshannaik did you try passing the map values are arguments and take the
measurements? Based on your earlier results it appeared that PersistentMap
lookup was causing the hit (I still think it could very well be due to the
profiler overhead)
Here are the changes I made -
https://github.com/arunmahadevan/storm/commit/7eae5ec9f63cee82c49980a3bedf5f0dfe4e3a8d
. I would like see how it affects your profiling.
I don't think a 0.4 to 0.5 % increase in throughput should be a reason to
completely disable a feature. And spouts that emit tuples in a tight loop would
not be a very common use case whatsoever. I am for documenting this feature so
that the users can adjust the config values based on their needs rather than
turning it off.
---
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.
---
