Skip to content

Conversation

@volyrique
Copy link
Contributor

Also, revert the recent experimental changes related to wrk that are meant to diagnose the performance regression present since the first continuous benchmarking run that includes PR #7783. Since neither #7832 nor #7852 lead to any significant differences in the results, the wrk update in #7783 is apparently a red herring, and the real culprit is elsewhere (most probably OS kernel updates in the host environment).

@volyrique
Copy link
Contributor Author

@nbrady-techempower Even though most of the checks were expected to fail, I went through the output to verify that there were no surprises. Most of the failures were in implementations that also had issues during the last continuous benchmarking run (i.e. were expected to fail), but there were some that were caused by unrelated problems - for example, many PHP and Rust implementations errored out with a no space left on device message.

@NateBrady23
Copy link
Member

Thank you @volyrique!!

@NateBrady23 NateBrady23 merged commit 5ea594e into TechEmpower:master Jan 26, 2023
@volyrique volyrique deleted the postgresql branch January 26, 2023 19:34
@sebastienros
Copy link
Contributor

I noticed a huge improvement on Multi Queries and Data Updates benchmarks today on ASP.NET benchmarks. And I was able to confirm it was due to this change. 15 to 20 % RPS increase.

franz1981 pushed a commit to franz1981/FrameworkBenchmarks that referenced this pull request Jun 23, 2023
…er#7877)

* Update MongoDB to version 6.0 and PostgreSQL to version 15

* Revert "Revert wrk to version 4.1.0 (TechEmpower#7852)"

This reverts commit 2e2afb8.

* Revert "Revert to using Ubuntu 18.04 for the wrk image (TechEmpower#7832)"

This reverts commit b14756a.

* Clean up several toolset Dockerfiles
Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment

Projects

None yet

Development

Successfully merging this pull request may close these issues.

3 participants