This ticket came out of the devprod Q4 retro

We've been making use of the perf cluster as part of the MSK load testing. This has been a great resource, but the perf cluster itself has raised some question related to process and management. Examples of some questions:

• Who manages the cluster and is there coverage for the cluster through US business hours?
• How are we to make changes to apps on the cluster, for example during load testing. We've been asked to contact service owners when there are changes, but we don't know who's actively testing, what we might break for people, etc
• Lack of knowledge on our team. Gabor seems to know the most about the perf cluster but there's little knowledge elsewhere

If we're going to make use of the perf cluster in future initiatives (and we should) then we need better guidance around how to use it, who to engage, and how to work on it without causing problems for others.