Test Groups

Test Groups allow you to organize and track tests within a single stage as independent sub-suites. This is the primary solution for supporting modern, parallelized CI pipelines where multiple test jobs run simultaneously.

Why It Is Useful

In complex pipelines, a single "Functional Test" stage might be split across 20 parallel agents. Without Test Groups:

• SeaLights would view all 20 jobs as one single entity.

• Recommendations might become inaccurate if agents are testing different builds at the same time.

• Savings data would be aggregated, making it hard to see which specific suite is performing efficiently.

Detailed Functionality

• Independent Tracking: Test Optimization recommendations are calculated separately for each group based on its specific execution history.

• Parallel Resilience: Different groups can run against different builds simultaneously. One group might be testing Build A while another has already moved to Build B; Test Groups ensure each receives the correct guidance.

• Granular Analytics: You can view execution data and optimization savings specifically for one group, rather than just at the stage level.

Implementation

When executing your tests, you provide two identifiers:

1. Test Stage: testStage The standard name for the testing phase (e.g., Integration_Tests).

2. Test Group ID: testGroupID A unique, meaningful name for the specific sub-suite (e.g., Payment_Service_Parallel_1).

Viewing Results

To see the impact of a specific group:

1. Navigate to Test Optimization → Savings Breakdown.

2. Use the Test Group filter to select the specific ID you wish to analyze.

Note: Test Groups affect Test Optimization logic only. Code coverage is still aggregated and displayed at the Test Stage level.