Optimizing Rails CI with parallel_tests

In our previous blog post on speeding up test suites with parallel_tests, we explored how the gem can significantly speed up test suites locally. Now let’s take it further — implementing parallel_tests in a CI environment.

Why Parallel Tests in CI?

Running tests in parallel in CI provides three major benefits:

Faster feedback — developers get results sooner
Scalability — handles growing test suites efficiently
Cost efficiency — better utilization of CI resources

The Single Job Approach

The single job approach with GitHub Actions is the most straightforward way to get started. It requires:

Multiple databases using rake parallel:setup
Utilization of VM resources
Simplified setup compared to matrix strategies

Step-by-Step Setup

1. Add parallel_tests to your Gemfile

group :test do
  gem 'parallel_tests'
end

2. Configure database.yml

Your test database configuration needs to support multiple databases:

test:
  database: myapp_test<%= ENV['TEST_ENV_NUMBER'] %>

3. GitHub Actions Configuration

Set up your workflow with PostgreSQL service, Ruby setup, bundler caching, and parallel test execution. Use rake parallel:setup to create multiple test databases (e.g., myapp_test1, myapp_test2).

4. Run parallel tests

bundle exec rake parallel:spec

Potential Challenges

Resource constraints — set PARALLEL_TEST_PROCESSORS to 80-90% of available cores
Database conflicts — solved by parallel:setup creating isolated databases
Over-provisioning processes — too many processes causes context switching overhead

Note that GitHub’s free plan provides only 2 cores, so adjust your parallelism accordingly.

Conclusion

With the right configuration, parallel_tests can dramatically reduce your CI pipeline duration, giving your team faster feedback and more productive development cycles.