[opminion]

A topic usually left out in scaling discussions is: how much can one predict? Or is it mostly trial and error? Is it mostly about good "reactive" engineering, would it have benefited from good mathematical modeling?