I’m new to Prefect (and new to cloud in general). I want to understand deeply how flow.serve and flow.deploy work. In the latter case we specify a workpool on which all computations get carried out, however, in the former case we don’t.
I understand that workpools aren’t persistant resources but an image running on a different machine everytime (correct me if I’m wrong).
Could anyone explain then what actually happens behind the scenes when you serve the flow and not deploy it, you would still need to provision resources for it, then what’s the difference, and what is the value you can get from one over the other?