Skip to content
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

TBD: Setup an continuously performance checking #71

Open
khmarbaise opened this issue Jun 16, 2024 · 3 comments
Open

TBD: Setup an continuously performance checking #71

khmarbaise opened this issue Jun 16, 2024 · 3 comments
Assignees
Labels
description missing/incomplete Size: M Esitmated size of the issue (S,M,L,XL)

Comments

@khmarbaise
Copy link
Collaborator

khmarbaise commented Jun 16, 2024

Goal

  • Run a performance analysing on a regular basis to see performance degrading etc. happens
  • It is inteded to identify problems during the development fixing of issue which might have influences on the performance of Maven itself.

Non-Goals

It is not a goal to

  • make a comparison with any other build tool.
  • ??

TODO

  • Produce automatic graphical informations about performance results
    • Different Maven Versions
    • Different JDK Versions
    • Memory consumptions
    • Using cache extension

Requirements for measurements

  • Select appropriate project(s) as foundations (some real life projects)
  • Create a number of synthetic generated project(s)
  • Define reference machine setup's
    • We need to have appropriate hardware to run those performance testing
    • My opinion is that it's can't be run on GitHub actions etc. because you will get all the time a different machine which means your measurements will never be reproducible.

Idea

  • Use a library to show the result of the measurements (for example echarts)

Reference

TBD more details...

@khmarbaise khmarbaise self-assigned this Jun 16, 2024
@delanym
Copy link

delanym commented Jun 18, 2024

Would this include some comparison with Gradle?

@khmarbaise
Copy link
Collaborator Author

Would this include some comparison with Gradle?

No I would not do that...

@rmannibucau
Copy link
Collaborator

Guess we should limit ourselves to ensuring we don't have regression so likely a few project kinds (single plain jar module - commons like, multimodule etc) and have a way to see the evolution with versions (ideally snapshot) to prevent any regression like we got recently.
Indeed it should be done with a single jdk for multiple maven version.

Having tests with multiple JDK for the same maven version can be worth it but since at some point they all converge I consider it as minor compared to identifying regressions in our codebase.

Hope it makes sense.

Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment
Labels
description missing/incomplete Size: M Esitmated size of the issue (S,M,L,XL)
Projects
None yet
Development

No branches or pull requests

4 participants