By name: integration-testing-performance description: Integration testing performance optimization, test parallelization, cleanup strategies, and CI/CD categorization patterns. Use when optimizing test execution speed, managing test data, or structuring tests for automated pipelines.
Integration Testing Performance and Maintainability
Use when speeding up integration suites, keeping data isolated/clean, and tailoring what runs in CI/CD stages.
When to use
- Parallelizing integration tests with isolated tenants/DbContexts
- Adding cleanup for shared containers or fixtures
- Tagging tests for selective CI/CD execution (PR vs main vs nightly vs deploy)
- Tracking slow/memory-heavy tests and optimizing data setup
Core principles
- Parallel by default with unique tenant IDs and fresh scopes per test
- Clean up created data (disposable scopes first; manual cleanup as fallback)
- Trait tests to slice by category/speed/feature/environment; align pipeline filters
- Monitor execution time and memory; fail fast on regressions
- Bulk-create data when needed; share expensive fixtures sparingly
Resources
- Parallelization: patterns/parallelization.md
- Cleanup: patterns/cleanup.md
- Categorization and CI: patterns/categorization-ci.md
- Performance monitoring: patterns/monitoring.md
- Test data management: patterns/test-data.md
Default locations
- Integration tests: tests/GloboTicket.IntegrationTests
- Shared fixtures/test data helpers: tests/GloboTicket.IntegrationTests/Fixtures or Helpers
- CI filters: pipeline scripts or dotnet test arguments in scripts/bash|powershell
Validation checklist
- Parallelization enabled in csproj and safe (isolated tenant IDs, no shared state)
- Cleanup runs (disposable scopes or explicit) to prevent cross-test pollution
- Traits applied consistently; pipeline filters match categories/speed
- Long-running tests have timeouts; memory checks guard bulk ops where relevant
- Bulk data creation uses single round trips; shared fixtures dispose correctly