Absolutely. I'm postulating a very conservative lower bound here.
(And, obviously, a very rough approximation of the actual history. If we could see Twitter's Git repository presumably we'd find that the service actually went through dozens, perhaps hundreds, of performance tweaks as it grew.)
Comments
> But it was up to the task of running Twitter for a few thousand users, […]
(Tens of thousands, actually, perhaps hundreds?)
Absolutely. I'm postulating a very conservative lower bound here.
(And, obviously, a very rough approximation of the actual history. If we could see Twitter's Git repository presumably we'd find that the service actually went through dozens, perhaps hundreds, of performance tweaks as it grew.)