| 2 Mar 2024 |
 raitobezarius | In reply to @nh2:matrix.org
Theoretically it does not matter so much. You can buy the HW, ship it to hetzner, and pay them 21 EUR for 15 minutes "Remote hands"
https://docs.hetzner.com/robot/colocation/faq/#can-hetzner-online-install-my-hardware Yeah, I guess | 02:00:24 |
| raitobezarius | I just don't know if we can afford or if we want to go for Hetzner Colo vs. Hetzner managed hardware if we go for a Hetzner service | 02:00:42 |
| raitobezarius | All of that is very fuzzy in my mind and dependent on current infrastructure team decisions / will | 02:00:54 |
| raitobezarius | So I would rather avoid pushing on them anything | 02:01:00 |
| raitobezarius | Also why I build my own solution on my side | 02:01:10 |
| raitobezarius | Overall, what I believe is that if someone comes in and drops 50K on the Foundation for the objective of duplicating the binary cache to save it, it's a solved problem | 02:01:46 |
| raitobezarius | Under that amount, you are kinda stuck having to ask the Foundation to consider taking the risk | 02:01:57 |
| raitobezarius | And it's not clear that it's a right choice or not | 02:02:04 |
 nh2 | For your colo, does it offer separate datacenters that isolated failure domains against fire? | 02:02:15 |
| raitobezarius | vs. garbage collecting and giving up on that data and preparing for the next set of data | 02:02:17 |
| raitobezarius | In reply to @nh2:matrix.org
For your colo, does it offer separate datacenters that isolated failure domains against fire? Nope | 02:02:27 |
| raitobezarius | If that was a goal, I would just send things in Glacier that I really care about, e.g. FODs | 02:02:56 |
| raitobezarius | In cache.nixos.org, there's cache :) | 02:03:02 |
| raitobezarius | Actually, cache.nixos.org was unavailable a lot of times | 02:03:20 |
| raitobezarius | But no one notice it in general | 02:03:23 |
| raitobezarius | Because a CDN is in front of it so… | 02:03:36 |
| nh2 | I mean in Hetzner it's a standard setup to setup servers different data centers to ensure proper HA | 02:04:32 |
| raitobezarius | Yes, yes, if we have the means and money | 02:04:49 |
| raitobezarius | Let's go for multi-site HA | 02:04:53 |
| raitobezarius | But again, from my PoV (maybe wrong?), short of sponsoring from hardware manufacturers or companies, it's hard to support those costs upfront | 02:06:06 |
| raitobezarius | Let's say, if HW manufacturers said: here's disks, here's servers (cheap in the context anyhow), and we could just pay colo, that'd be interesting | 02:06:30 |
| raitobezarius | etc. | 02:06:33 |
| nh2 | This is not what I mean. I mean that if you buy the hardware for a colo, it costs a fixed price. You could put it into your colo in France, or in Hetzner, where you would get proper HA. | 02:06:33 |
| raitobezarius | In reply to @nh2:matrix.org
This is not what I mean. I mean that if you buy the hardware for a colo, it costs a fixed price. You could put it into your colo in France, or in Hetzner, where you would get proper HA. This, I don't disagree | 02:06:56 |
| raitobezarius | But what's your point regarding that? | 02:07:07 |
| nh2 | My point is: Hetzner is currently the cheapest $/TB hardware rental on the planet. The only cheaper way (long term) is owning the hardware and putting it into colo.
So I'm just pointing out, if you're considering colo, colo at Hetzner might be the way to go for the absolute cheapest and still reliable. | 02:08:52 |
| raitobezarius | I'm not extremely familiar with Hetzner colocation | 02:10:31 |
| raitobezarius | but reading from that https://www.hetzner.com/colocation/ | 02:10:35 |
| raitobezarius | assuming 1kW (constant) over one year, assuming 2 locations in Finland, at the announced electricity rates, this is something like 2×(119 + 0.2618€/kWh1kW1year) = 4827,87 € | 02:12:28 |
| raitobezarius | * assuming 1kW (constant) over one year, assuming 2 locations in Finland, at the announced electricity rates, this is something like 2×(119 + 0.2618€/kWh*1kW*1year) = 4827,87 € | 02:12:38 |
