-
Notifications
You must be signed in to change notification settings - Fork 141
New issue
Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.
By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.
Already on GitHub? Sign in to your account
Make benchmarks more accurate #1484
base: master
Are you sure you want to change the base?
Conversation
fcc5f2b
to
cf554d7
Compare
There was a problem hiding this comment.
Choose a reason for hiding this comment
The reason will be displayed to describe this comment to others. Learn more.
If I'm understanding it right, this is applied to all benchmarks so doesn't give zio-kafka an unfair advantage, just a more realistic comparison, right?
Yes, that is correct. |
a4eab98
to
2656123
Compare
... by adding a 5ms delay in each call to the underlying java consumer. This solves the problem that the embedded kafka broker can respond virtually immediately, while in reality there is almost always a network hop. Since the calls of the underlying java consumer do not correspond 1:1 to network traffic, this is a crude approximation. Nevertheless, we expect that redundant consumer calls will become visible in the benchmarks. Also: remove a duplicate benchmark.
2286df1
to
3c65ec1
Compare
... by adding a 5ms delay in each call to the underlying java consumer. This solves the problem that the embedded kafka broker can respond virtually immediately, while in reality there is almost always a network hop.
Since the calls of the underlying java consumer do not correspond 1:1 to network traffic, this is a crude approximation. Nevertheless, we expect that redundant consumer calls will become visible in the benchmarks.
Also: remove a duplicate benchmark.