If the numbers can be satisfied by a Postgres then thats the correct answer. The interviewers fucked up, because they sized the problem wrongly.
This is the same issue that was prevalent when the industry switched from HDD to SSD: some system design questions suddenly became trivial, because the IOPS went up by magnitudes. This is not a failure of the interviewees, who correctly went with the times, but a failure of the interviewers.
> then that's the correct answer.
Obviously. But they don't care about that answer, they care about the design of a new system so I'm saying the interviewer is dumb for not just asking directly for the solution he wants to see (while also giving credit for the original right answer).
The interviewee is dumb (if he wants the job) to not know this is what they want to hear, so he can just say "I really do think Postgres would satisfy the requirements here, but as an alternative a custom solution would look like.."
What is the point of not demonstrating your own well rounded knowledge when given the opportunity? Is it not obvious what they want to hear? Then why not tell them, again given that he did want the job
The correct answer is “Postgres would handle it, but if it needed to scale even higher, I’d…”
The point of a system design interview is to have a discussion that examines possibilities and tradeoffs.
What kinds of system design questions got destroyed?