Science fiction is fiction
https://www.youtube.com/watch?v=H_SaKXM82yg&list=UU9rJrMVgcXTfa8xuMnbhAEA - video
https://pivottoai.libsyn.com/20251201-ai-data-centres-in-space-why-dcs-in-space-cant-work - podcast
time: 9 min 27 sec
Data centers in space are a tool. You have to know when and how to use them. I’m not saying they’re completely useless, but most people do not understand how to avoid the difficult and expensive orbital logistics, power and cooling issues, radiation problems or the slow and complicated networking (unlike me, of course people like me know how to avoid them). Obviously it’s ludicrous to suggest space station server farms don’t have their uses and I’m not the kind of luddite saying nobody should ever be putting data centers in space, but right now they should really only be used together with terrestrial data centers and not relied on exclusively. That said, it’s still early days and we will inevitably be seeing a lot more compute in the orbit.
My poe detection wasn’t sure until the last sentence used the “still early” and “inevitably” lines. Nice.
I get what you’re saying.
I hate when they make us untangle these linguistic knots. I think this is a motte-and-bailey1 situation, like how the oppo respond to AI (LLM) criticism by implying we don’t like AI (ML i.e. the old thing that works).
They likewise want datacenters (Big Hot Steaming Shitbox of GPUs…IN SPACE) not datacenters (radiation hardened, multiply-redundant, low-power industrial CPU clusters…IN SPACE)
1 ugh sorry having a brain fart and cant think of a better term
Edit: I am reliably informed that I let one go over my head. Apologies!
In my defense, I would be entirely unsurprised if you turned out to be an expert in this stuff lol
I was just riffing on the AI “moderate” talking points. Building a data center in space is prima facie ludicrously stupid and you would need an extremely unusual justification to even consider it. I was pretending to act like a moron who blindly accepts there’s probably a serious reason why they make sense just because some dumbass hype man said so.
The place for space data centers is in manned space stations to support the work being done there.
All the venture capitalists who got scammed into funding this:
space is cold
“Data centers on hard mode” is such a good way to put it.
Space is cold; but if something is hot in space, doesn’t it also need to, like… Push the heat away somehow? No air, no wind, heat just stays there. 🤷♂️
Yup, they can only dump heat through infrared radiation. The giant white fins on the ISS are radiators for waste heat.
I think there is another way to dump wase heat. Use it to warm up something that has a lot of heat capacity, and then eject that. Not practical and a huge waste (I also got this from science fiction), so you know they are going to try it. Gonna suck for us groundies.
“Run, the Muxk servers are overhead, it is raining molten tin!”
Why not submerge them in a river for far less money.
Microsoft had an underwater datacenter-in-a-container test a while back.
He touches on that in the blog post, and gives us this link: https://www.itpro.com/infrastructure/data-centres/microsoft-scrapped-its-project-natick-underwater-data-center-trial-heres-why-it-was-never-going-to-work
Didn’t Meta, then Facebook, open a datacenter somewhere in the Arctic circle and leave it exposed to the environmental cold? Until the planet runs out of cold, an eventuality whose impending rapidity I’d rather not ponder, that seems a viable approach.
“How do we cool these things?”
“idk, boil the oceans?”
I’m surprised lagering caves, except for servers instead of beer hasn’t come up yet.
Underwater datacenters make cooling very effective and maintenance nearly impossible, so you have to treat the container data centers essentially disposable. That’s only viable with economy of scale big enough to be an xkcd comic punchline. I guess Microsoft found that even they are not quite there yet. Also most computers don’t tolerate seawater quite as well as they tolerate air.
Pretty sure space datacenters are also going to be even more disposable.
Oh yea absolutely. Underwater datacenters have one upside (cooling) and massive downsides (everything else, more or less). Space datacenters trade that upside into yet another downside, make the downsides even bigger and add a few extra downsides for good measure.
Cooling in space is an absolute arse. Space is an excellent insulator for heat. That’s why a thermos works. In space, thermal management is job number one. All you can use is radiators. Getting rid of your 200 kilowatts will need about 500 square metres.
To drive home how easy this is to work out, the Codex for the Mass Effect series1 explicitly points out that radiation is the only way to cool off in space, and goes into detail on how in-universe spaceships (civilian and military) deal with heat buildup.
BioWare did their homework on this shit for a series of sci-fi RPGs which started in the early days of the Xbox 360 and the PS3. That the startup bros, tech co’s and billionaire CEOs pushing this have failed or refused to recognise this shit is goddamn negligence.
So space is a bit hard. A lot of the sci-fi guys suggest oceans! We’ll put the data centres underwater and cooling will be great!
The only way I see that idea working is if humanity works out underwater cities (e.g. Rapture from the original Bioshock) first. That’d make the issue of maintenance easier to deal with, even if getting new parts from the surface would remain a PITA.
1 Specifically “Starships: Heat Management”, under Ships and Vehicles, in the Secondary Codex"
The tabletop game Attack Vector: Tactical from 2004 also models the need for radiators and risk of overheating.
Science fiction from the 20th century tended to ignore cooling and cosmic rays (the one exception I can recall is Jerry Pournelle’s superscience Langston Field) and we know these guys are not up to date even on pop culture or good at reading.
