E2E: Service performace for gitlab jobs

​🆕​ Task

Affected app variants: All

Summary

Investigate service performace for gitlab jobs, and decide if this approach would do what we want.

Background

Check out how to improve the performance "services" defined in the jobs. If this works out, it can be great as we wouldn't have to do anything else. However, one problem with this approach is that services are spawned per job. That means that the service image gets started multiple time at the same time, and this is probably the reason why everything gets some much slower.

Acceptance Criteria

• Running experiment with improved configs or decision that shows that this is not a proper approach

Results

I have performed different tests on the pipeline including multiple jobs and their services and I have found out that it is not possible to have global services for multiple jobs, which results in having a bunch of services running per job. We would need to upgrade our runners in order to really see a benefits. I will continue with the second part of the experiment which is spawning a global temporal service.