Github user mtaylor commented on the pull request:
https://github.com/apache/activemq-artemis/pull/227#issuecomment-99431498
The test consisted of sending a million empty messages 5x, taking average
time. The test is in no means 100% accurate as it was running on my dev laptop
with other services running.
The results show that running without interceptors, the test ran 700ms per
million messages slower.
This is obviously the lack of precision in the test as we would expect at
least this to be equal or faster.
If we believe that this is a serious concern then we can run more accurate
benchmarks. However, I believe there is no significant over head in this
approach.
Results are as follows:
=====================
Without Interceptors - Produce and Consume same session/thread:
Iteration Time: 58670
Iteration Time: 56232
Iteration Time: 55591
Iteration Time: 55013
Iteration Time: 55985
Average Time: 56298
Produce only:
Iteration Time: 19518
Iteration Time: 18240
Iteration Time: 13952
Iteration Time: 14758
Iteration Time: 17506
Average Time: 16794
======================
With Interceptors - Produce and Consume same session/thread:
Iteration Time: 59663
Iteration Time: 55300
Iteration Time: 54737
Iteration Time: 55259
Iteration Time: 53029
Average Time: 55597
Producer only:
Iteration Time: 19566
Iteration Time: 16898
Iteration Time: 13981
Iteration Time: 15014
Iteration Time: 14881
Average Time: 16068
---
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.
---
