Set up a system for automatically running load tests after deploy to YT

[SandGrainOne]

Description

We want to track the performance of our solution and be able to quickly identify if changes to our code or new features introduce a degradation of the performance of platform.

One way to do this is to have bench mark load tests that can be run with each change / deploy to YT.

Consideration

Test results from a single test gives us little insights. We will need somewhere to persist and present the test results to easily identify if and when a change to the performance occurs.

Should the number of replicas match what we are running in production at the time or be scaled up to a maximum? And regardless of what we go for here, what is the best way to scale the environment up and down with minimal effort?

Tasks

• Identify relevant tests for the platform components tests suites
• Set up a pipeline (DevOps or GitHub) to trigger relevant tests
• Set up dashboard to present test results

- Automatisk kjøring av RF-0002 testene etter deploy til YT01
- Skalere automatisk opp før kjøring og ned etter kjøring.

Acceptance criteria

• Issues are created describing what tests are missing for the test suite to be complete
• There is a dashboard available to the team where the performance of the solution over time is available