News for hackers. Stuff that matters.
Re:Fly.io Postgres cluster down for 3 days, no word from the... by ushakov on Friday July 21, 2023 @04:43AM (#36811511)
Me and my partner have paid a visit to their datacenter in Nüremberg. The answer is efficiency. They get more processing power than the other providers for the energy they have to put in
Re:Fly.io Postgres cluster down for 3 days, no word from the... by arrowsmith on Friday July 21, 2023 @04:59AM (#36811622)
What do they do that makes them more efficient?
Re:Fly.io Postgres cluster down for 3 days, no word from the... by abwizz on Friday July 21, 2023 @05:29AM (#36811798)
i'll guess they pick optimized components for it.

like the longtime workhorse was a high performance skylake desktop cpu w/o ecc ram

Re:Fly.io Postgres cluster down for 3 days, no word from the... by ushakov on Friday July 21, 2023 @09:24AM (#36813464)
The secret is in the cooling system. They have individual cooling systems for each server. Less heat = longer sustained loads
Re:Fly.io Postgres cluster down for 3 days, no word from the... by abwizz on Saturday July 22, 2023 @04:55AM (#36824593)
pardon my ingorance but i cannot quite see how cooling individual machines vs. the hole rack or row makes a difference in total heat production per machine