Vacuum does not vent heat.
This has been a PSA.
Vacuum does not vent heat.
my Dyson gets pretty warm.
Begrudging upvote
Still kinda blows my mind that a ship could accelerate to 1c and still not be able to use cooling fins, aircraft on earth use it as part of the design process
no worries about the temperature if the ship collides with a spec of spacedust :)
No but hot things will just radiate it away, regardless of vacuum.
Yes, but the infrared radiators for a single megawatt are apx the size of 4 tennis courts. Until there’s a space elevator, that’s a tremendous amount of resources to get into orbit.
Yea, because the only resources in the entire universe are only on our planet. There’s absolutely none out there where there’s little to no gravity well to fight against right?
Sure, but we’re not doing anything to mine or manufacture in space.
And we can’t, until we do.
True enough.
I eagerly await OP’s space mining operations, which Elon has promised to start doing for a decade but simply hasn’t because he just didn’t wanna.
Apparently there are a lot of concerns, of which, one is venting heat. Also the change of heat as a satellite dips in and out of sunlight.
I didn’t think about radiation, which I should have since it’s a not-small problem for human astronauts.
Heavy CMEs are going to be the AI equivalent of Purple Haze LSD.
Just confirming what I’ve suspected: the point of the space datacenter thing is to keep the tools of oppression out of reach of angry mobs with torches and pitchforks.
Data center in space: untouchable
Strictly necessary ground radio station to get any usable amount of data to and from said data center: still flammable
And I don’t know if the space datacenter is quite as untouchable as he thinks…
It will obviously be in a very low orbit, because that’s the easiest to reach and construct…
-
Maybe it would be within reach of some fairly advanced amateur rocketry? We don’t have to reach a stable orbit, which makes things much easier. Just shoot it straight up, with much less velocity needed than for orbit. In fact, the speed difference between our rocket and his orbital datacenter will be very useful – that speed difference will lead to a very energetic impact.
-
Maybe it could be vulnerable to some directed energy weapons? Very strong lasers, highly directional microwave beams, that sort of thing.
-
There’s always hacking. For extra irony bonus, use Space Grok to vibe code the tools you need to hack its orbital control systems. Then you can de-orbit it or collide it with other objects in orbit.
It would probably be easier to launch shrapnel into their orbit than try to hit one, and let them begin to cascade into each other. The dishes are so low (or at least starlink is), that there’d be no real kessler risk since it’d all fall back down to earth in 5 years or less.
The following is a reference and spoiler for Metal Gear Solid 4: Guns of the Patriots
spoiler
What we need to do is hijack a submersible fortress, steal the prototype railgun designed for stealth-launching nuclear warheads, and mount the rail gun on the fortress so you can guard it with your battle-harem of cyborg soldier women. Bonus points for getting your arm chopped off by a ninja, replacing it with a transplant, getting possessed by the donor’s ghost because you were unknowingly the child of a medium, but then getting the transplant arm replaced by a prosthetic and hypnotizing yourself into thinking you’re still possessed because reasons… Oh yeah, point the railgun at the orbital data center while all this is going on to ragebait the clone sibling of the arm donor I guess!
Easy!
Man, those games got wild… I mean, they started pretty wild. But yeah.
Indeed. I had to remind people that this was the actual third/fourth act of a flagship title and I was among the many people who took it very seriously because it “changed gaming forever, it was art!”
I mean, I still love Metal Gear Solid, but we have to be able to see it with some objectivity
We have had missles that you can lob from a high altitude aircraft, and hit a satellite, since the 80s.
[…] that speed difference will lead to a very energetic impact. […] Then you can […] collide it with other objects in orbit.
How to make space worse and everyone hate you - 101
Well, maybe don’t fuck over most of the population so much that they see no progress as preferable to progress that only benefits the oppressors.
It will obviously be in a very low orbit, because that’s the easiest to reach and construct…
Well, for a bunch of reasons, these space data centers probably won’t be at very low orbits. First and foremost, very low orbits like the ones starlink uses experience a lot of atmospheric drag. With the huge solar arrays these data centers will need, that high drag environment is a killer. Going just a bit farther out would significantly reduce that drag. Also, latency isn’t a huge concern with space data centers, so very low orbits aren’t required. Lastly, since those low orbits are so highly valuable and contested, it just doesn’t make sense to put something there that doesn’t need to be there.
TLDR: Space data center orbits are likely to be somewhere between 500 - 1000 km altitudes.
And yes, that’s in untouchable range for anything but advanced militaries.
What it’s not untouchable to is legislation. That’s the practical way to fight it.
I really hope they don’t go into the 700-1000km range with something of this scale if it actually happens as thats where a real risk of debris staying in orbit for hundreds or thousands of years starts to happen if something catastrophic happens to the constellation.
China blew up a satellite in a test in the 800km area and it’ll be there for hundreds of years.
Yeah, that’s a valid point. And yeah, that’s totally where they’re gonna put them. They’re gonna be huge, delicate targets…
I’m generally pro space, I think all the massive advances in space flight in these last couple of decades has been truly amazing. That said, I wish I had a more positive way to spin this, but yeah, that’s a serious concern. That’s bad.
It won’t happen, but if globally we could agree to put a moratorium on large constellations in those orbits until we have more experience with constellations in the really low orbits, that would at least help and give us more time to learn. Without everyone agreeing though, someone else will just be like, MINE and fill it anyway.
Edit: like lets see what happens as starlink becomes even bigger and moves into 10s of thousands, and kupier appears etc before moving up higher.
Very low orbits are less of a concern, though. Because if the worst happens, they and the vast majority of their debris will be slowed by atmospheric drag and fall relatively quickly.
The latter two would be way easier.
You’d think… But a laser or microwave emitter that big would also be quite an undertaking.
Honestly, hacking seems the most likely option to succeed on a small scale.
Oh absolutely. Waiting and pouncing on some vulnerability would be the safe bet. Maybe blind their sensors or command and control uplink radios with lasers or microwaves, not damage them directly.
Can we please not destroy objects in orbit? It’s already crowded up there without untrackable debris. Deorbiting is much safer and probably easier, considering the precision needed to hit anthying, especially because those satellites would have station keeping thrusters to evade other satellites.
Great way to enforce a moratorium on new large-scale projects in LEO for a few years.
-
You could burn the ground stations down, but they’re all around the world, and all it takes is 1 to route the data to via the inter satellite laser links. You’d definitely add latency though taking out ground stations near you.
Good thing the whole idea makes no sense, because repairs are impossible, you can only vent heat in infrared, you have to use expensive rockets to get them up there, and if they’re in LEO, they’ll fall back down to earth. Oh, and there’s more likelyhood that bits get flipped from all of the high energy particles up there.
It’ll just make LLMs at scale even more infeasible.
To run one rack of GPUs they will be about 100 square meters of solar panels and 30 square meters of radiators. That’s about 130 kilowatts of compute power. To get a gigawatt multiply that by 6,000. Musk is claiming he wants to send terawatts up to there.
The numbers didnt make sense for his tunnel thing either, but thats because he has never been an engineer but a hype-man who manipulates governments and markets for proffit.
The stats you listed are close to what 1 satellite would be if memory serves.
But when he talks about sending terrawatts up there, he’s talking about having a moon base manufacturing them and launching them into orbit with a mass acceleator, so if we’re doing that… ramping up some solar/radiator manufacturing capacity isn’t as far fetched. Were fully into sci-fi land at that point.
Just FYI the rockets won’t be expensive if this plan ends up working.
The only way this works beyond a tech demo, is if Starship works.
If Starship works as planned (big if), the marginal cost to orbit will round down to the cost of fuel to launch over the lifetime of the rocket.
Starship being fully reusable and working is going to be a mind boggling change in what access to space looks like and means.
Yeah it’s tricky, given the pushback people are showing against datacenter construction, the issues around power, the vulnerability to attack… The space data center idea is starting to become more attractive to big tech over time, not less.
Initially, everyone agreed that the numbers just don’t add up, there’s no way it could be economical. But then these terrestrial datacenters started getting a lot of bad press, everyone was concerned about the environmental impact, about the strain on the energy grid, about water usage. All of a sudden the space plan is seeming less crazy.
and if they’re in LEO, they’ll fall back down to earth. Oh, and there’s more likelyhood that bits get flipped from all of the high energy particles up there.
You’re thinking about the right things here, these are major issues! But, they can largely avoid serious problems here. Over 500 km, there’s a lot less drag and these aren’t just going to fall out of the sky. And under 1000 km the earth’s magnetic field still mostly protects them from high energy particles. They will still definitely be more exposed to high energy particles causing random bit flips, but it’s going to be in a totally manageable range.
Giant AI datacenters will truly never be an efficient or wise use of resources, at least in my opinion, it’s just a largely stupid endeavor. But for better or worse, I expect we will be seeing some of these space based data centers in the near future. I guess we’ll just have to wait and see how economically sustainable they are…
Bad press doesn’t mean they can start losing even more money on space data centers (which are going to be incredibly inefficient and costly). They already do not make a profit, they’re all deep in the red.
Drag is not the only problem (even though these would definitely be LEO, it is just even more expensive to not have them in LEO and it increases latency), nor are bit flips. The biggest issue will continue to be heat. If you haven’t been to a data center before, you may be surprised at how much heat they generate. I’ve only been to non-LLM data centers, and the heat in the warm aisles was unbearable.
And that’s without the sun heating the exterior to incredible amounts. You cannot use contact to dump heat into space.
You might also be surprised at how much e-waste data centers generate, especially LLM-geared data centers. They obliterate hardware, even with optimal cooling, and that hardware requires constant replacement. How are they going to manage that?
Also, unless they use nuclear (which would drastically increase costs and add heat) they’re going to need gigantic solar arrays, much bigger than the IIS’ array.
you can only vent heat in infrared
Yeah, when you heat something, it emits infrared. Kind of the nature of hot things.
I think CorrectAlias meant that heat transfer by direct contact, which is how we cool things in general, is not available, making cooling a datacenter cluster in space actual science fiction.
This is exactly what I meant, thanks.
@phutatorius @technology So, a high-efficiency thermoelectric generator, powered by #datacenter heat directly, might be a win-win. I don’t understand why the current (jk) #technology efficiency is still so low. Surely someone must be working on this.
you still need thermal radiators to create the temperature differential.
There’s always lasers…
Instead of shipping an entire data center to space, let’s ship a single Elon instead.
This leaves more room for fish to thrive.
Please, waste your money on something that needs several football fields big heat radiators per gigawatt. And cannot replace a hard drive without launching a technician into space.
Easy, just tell the AI to change the harddrive!
Elon should do I it. Really. Either we get to see what noone understands yet: How economically viable server cooling at scale in space looks like or we see Elon faceplant financially yet again.
Either way, we win.
I want Elon Musk in space, aimlessly drifting around.
I would chip in $25 to send him to space, unless it was a two-way ticket.
Hilarious prank though. Send him into space, and then “accidentally” forget some of the heatshield.
Save money on weight, just send his head up.
I can live with that compromise. Kg price for sending shit to space is high.
I mean, we could try but I don’t know how we’d fit it in there
lol. I don’t wish him any harm, I just want him far from those he might continue to harm.
He will do absolutely anything to not spend money on making the world he lives in a better place for all. So build a data center in space, just add a little apartment for yourself. Imagine looking down on the whole world! What power!
He’s counting on Mars. Which isn’t insane. He should go. We’re right behind him.
I’m still waiting on how he’s going to manage CMEs during spaceflight.
NASA timed its manned high-space missions to low-ebb periods, and there was still some risk of astronauts being baked to a crispy golden brown.
Getting to mars is nine months, rather than a few days to the moon. CMEs be coming.
Is there somewhere we can donate to help make sure he
fries in a coronal mess ejectionlives the dream he’s always wanted?With our current tech, a Mars mission pretty much 100% certain a one-way trip.
Common elon, wheres your “move fast and break stuff” spirit? Lets get you suited up and on a rocket asap!
Imagine how rapidly he could innovate on a constant acceleration vector out of the solar system!
Just a reminder, while they speak of data centers in space, their proposed satellite is actually more like a single server rack in space.
It’s always misleading, you can’t even trust their choice of words.
Democrats immediately win three of the four seats on the Moon and sweep the Sky Council.
We can also launch Elmo into space and… Well, just forget about him, good riddance
lol let him waste his money on an impossible technology
He doesn’t waste his money on his companies. He wastes tax payer funding on them. And when that started drying up, he took it to the stock market, convinced the exchange to fast track adding his company to being eligible for indexes, and then wasted our 401ks on it. He’s the ultimate con man.
Yes but that only lasted weeks before collapsing. The end is coming, he is now trying to roll Twitter into spacex. It’s where his real riches are. When it eventually goes, so does his money
That’s what he does. When one company starts having issues, his other company will bail it out or outright buy it. It’s a constant shell game.
I think my point is that his wealth is based on the game and like anyone who passes bad checks between accounts, the reckoning eventually catches up. He’s going to lose big over the next few years.
I’m hoping for a push by the people to pressure the Democrats to defund Musk. All his tech is shit and hurts people.
With Trump, he got no-bids.
Unfortunately there are still a ton of people that think he is some genius engineer and visionary. I would love for him to get cut off, I just don’t have high hopes for it. Ideally he is cut off and sent to prison or sent back to South Africa.
Yes. The investment market is based entirely on vibes, which means Trump/Musk shysters thrive. That might be a good use for AI that doesn’t hallucinate: create a data-driven investment advisor that works.
While we’re at it, a data-driven c-suite manager that actually cares about productivity will make workplaces much less toxic. One can dream.
And kill us all by destroying the ozone layer again? No thank you.
I’m not sure what it has to do with the ozone, please explain.
One of the major environmental issues with launching mega-constellations is the environmental impact of their end-of-life deorbiting. As satellites reenter the atmosphere, they create aluminum oxide particulate that has a negative impact on the ozone layer.
Because, of course they do. We always have to find ways to destroy our beautiful planet, as if a new Earth-like planet is just a skip away.
Thanks
An utterly unworkable, absurdly costly idea. I’m sure the same investor-lemmings who buy heavily into IPOs which have no viable business model will fall for this too.
Underwater data centers are also not that feasible, but they are a lot more feasible than putting one in space.
Yep. We already have to deal with waste heat on the ISS, where is the monstrous heat from even a traditional DC to go?
There’s only a rack worth of compute on each satellite. There is nothing monstrous to deal with individually. The plans are to put up a very large quantity of these which will sum up to a large amount.
That probably makes training on them less effective, and that would likely remain a land based thing, but for inference, any given person probably doesn’t need more than 1 rack of compute to themselves, and if they do, they’d probably be using something land based.
Elmo will have to move to Floriderp.
kinda hard to do that when he invested so much time and money on blexas
A data center in international waters is immensely more practical in most regards, if legal jurisdiction is your complaint.
Great idea. And for each data center ship you can have a desalination ship for the cooling water. And both will burn bunker fuel, of course.
Would the water need to be desalinated if it’s only going to be cooling the equipment?
If you want the cooling system to survive, yes.
Ocean water is atrociously corrosive, prone to clogging tubes and short circuits much more easily than distilled water.
Exactly. You could of course construct a ship with a closed inner cooling cycle that then dumps the heat to the outside ocean through cooling fins or the like. But that would require developing new technology and building new ships. That would take time and cost money. We can’t possibly do that because line must go up now!
This isn’t a new idea. This is a problem that every nuclear sub deals with. From what I can gather, fresh water runs inside the reactor in two separate zones, and after powering the steam turbine goes into the condenser which is cooled by seawater.
However it works, this problem has been solved for decades.
I’m not saying it’s impossible or even difficult. But the fact is that currently no ships exist that implement this on the necessary scale. You don’t just build a completely new type of ship in an afternoon.
developing new technology
This isn’t as hard as you want to believe, nor would it require a brand new ship. If you want to do coastal designs, there are already some in place. If you want data sovereignty, the only real benefit of space data centers, a small cargo vessel with container data centers and one or more onboard generators could do this. Fueling it would be a challenge, but no more than all the issues with a space data center. As for time constraints, this would still be faster to deploy than any space-based solution.
None of this makes it a good idea, just not particularly novel, difficult, or time-consuming.
Good luck dumbfuck, tell me how easy it is to service, maintain and repair them when you inevitably overheat them, or don’t have the proper shielding so components burn out or degrade beyond useful parameters.
Mars is all red, maybe you want to restart those plans and take your flock of sheeple with you as soon as possible?
As much as I dislike Musk, he probably could tell you in detail. His company is not the only one looking into compute in space, either.
Yeah, other CEOs also fail basic physics.
Almost like Musk had to buy his physics diploma, eh?
Nah, it’s a grift. The Melon’s pitch straight up includes factually wrong physics claims. It is not easier, nor cheaper, to have data centers in space.
Actual scientists have done the math. Even disregarding maintenance for data corruption from radiation, and assuming the best near future tech. It would take 10 times an ISS equivalent structure to power and cool down a single server rack.
Musk promised he wants to get 1 Tera Watt of LLM processing power in orbit. That’s 10,000 times an ISS of solar panels and IR radiators for cooling. The enterprise is completely unrealistic.
The whole thing is just a con to get governments to funnel more money into his companies without having to deliver any actual results.
there’s no convection cooling in space, because vacuum is a very good thermal insulator. compute in space does not scale because of that alone. you would need massive (F/M)IR radiator arrays to get rid of the thermal energy.
And every single one of them is a grifter. It would be a lot easier to establish a permanent moon settlement (which can then build datacenters if they want to) than to provide meaningful amounts of compute in orbit.
