Skip to content

Conversation

@anmarchenko
Copy link
Member

@anmarchenko anmarchenko commented Jan 30, 2026

Summary

  • Changed the default value for --min-parallelism and --max-parallelism from 1 to the number of available CPUs (runtime.NumCPU())
  • This respects container CPU limits in cloud CI environments, so users get appropriate parallelization out of the box
  • The README already documented this as the intended behavior, this PR implements it

Motivation

Having parallelism set to 1 by default is not terribly useful for a parallel test runners: the number of available CPUs is more sensible default as we assume that we'll run tests in parallel on the current machine.

Test plan

  • All existing tests pass (make test)
  • Linter passes (make lint)
  • Tests that require single-runner behavior now explicitly set parallelism to 1

@anmarchenko anmarchenko requested a review from a team as a code owner January 30, 2026 11:44
@anmarchenko anmarchenko changed the title Set default parallelism to number of available CPUs [SDTEST-2703] Set default parallelism to number of available CPUs Jan 30, 2026
Copy link

@chatgpt-codex-connector chatgpt-codex-connector bot left a comment

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

💡 Codex Review

Here are some automated review suggestions for this pull request.

Reviewed commit: e01ce6880d

ℹ️ About Codex in GitHub

Your team has set up Codex to review pull requests in this repo. Reviews are triggered when you

  • Open a pull request for review
  • Mark a draft as ready
  • Comment "@codex review".

If Codex has suggestions, it will comment; otherwise it will react with 👍.

Codex can also answer questions or update the PR. Try commenting "@codex address that feedback".

When a user only sets --max-parallelism to a lower value (e.g., 1),
they should get that value, not the default min (which is now NumCPU).

Previously, if max < min, the code returned min, effectively ignoring
the user's cap. Now we clamp min to max instead, ensuring the user's
explicit max setting takes precedence.
@anmarchenko anmarchenko merged commit a6a91af into main Jan 30, 2026
3 checks passed
@anmarchenko anmarchenko deleted the anmarchenko/default_parallelism_is_number_of_cpus branch January 30, 2026 12:09
Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment

Labels

None yet

Projects

None yet

Development

Successfully merging this pull request may close these issues.

3 participants