Runner performance best practices?


We’re using EC2 instances as runners, this works well. But are there any recommendations for what types of instances to choose? I assume this largely depends on the type of jobs that the runner is doing?
Perhaps just try out some different types (t3, m2, etc)

If the runner is neither too slow, nor too expensive, then there’s no problem? My guess is the fine tuning begins to matter only once you need thousands of these instances at the same time, because that’s when things start to cost some serious money. As always, you’ll have to benchmark your needs. I imagine they’d have a control panel page dedicated to that somewhere.