Only stating the solar panel - radiator area ratio and not mentioning the actual area that a data center should have, is a sin of omission - and a great one. Also, hand-waving numerous engineering problems just by saying “we’ve solve harder problems before / you can come up with a few designs” is the last thing I want to see in anything that comes up in Hacker News.
Allow me to propose a modest alternative to space data centers, namely mountaintop data centers. This would consist of a container full of servers and GPUs and what else goes into a data center, a wind turbine for power and a communication module (say laser or microwave) for communicating with a base station with a fiber connection. This would be lifted on top of a mountain by a helicopter and bolted in place. Cooling would be provided by heat sinks exposed to the outside air. Some of the nodes could relay traffic from other nodes on remote mountain tops out of sight of the base station.
This scheme has many advantages over space data centers including launch costs, cooling, connection latency, servicability and ease of recycling.
I think we can't rule out the explanation that all the ideas of space data centers could be connected to a desire by some of finding additional applications for rockets that can transport stuff to space.
It also provides a post hoc rationale for rolling Elon's loss making businesses into SpaceX. The bull case for ODCs looks a lot like the bull case for space based solar power that Musk once called "the stupidest thing ever"...
That said, SpaceX aren't the only entity proposing ODCs, they're just the only ones promising they're going to make country-sized profits out of them...
I'm thinking if we send them out to catch the Voyager probe up, people might have time to write stuff themselves before they ask the computer to do it for them :)
Maintenance for a mountaintop data center only requires a team of skilled mountaineers. In space you'd need astronauts. It's at least an order of magnitude cheaper, perhaps two or three.
What advantage does a mountaintop have versus a more accessible earth-based location?
I suspect one of the motivations for space datacenters is to try to stay out of the reach of all jurisdictions so you Musk can start to run his companies as an autonomous state.
If the goal is to avoid various national jurisdictions, what about floating a barge out into international waters? Do it near the equator in the middle of the pacific to maximize the difficulty of humans getting there. Use the money saved by not needing to launch into orbit to purchase massive gun turrets to prevent piracy.
It is a comment on the absurdity of orbital data centers. Mountaintop data centers sound absurd, but are more feasible and efficient than orbital ones in nearly every aspect.
Space data centers are physically possible but don't make financial sense. The total cost of an orbital datacenter over five years is at least 2–3x that of a terrestrial one.
But those economics don't matter to SpaceX, because the main purpose of its orbital data centers is to create a use case for Starship. Starship has to fly frequently to iron out the kinks, encounter and fix rare (1/1000) failure situations, and optimize the launch cadence which pushes launch costs down. Plus Starship needs to fly a lot before it's ready for crewed flight. The long-term goal is a Starship optimized for crewed interplanetary travel. Orbital data centers are a payload that bring in some revenue, and provide a reason to launch constantly.
It's the same thing they did with Starlink to make Falcon 9 as reliable and rapidly reusable as it is.
There’s so much data center capacity being built all over the Earth. Thousands of large projects across US / China / Europe / Middle East. It would be astonishing if something that’s never been done before could be so cost-competitive immediately.
Starlink wasn’t the first time LEO communications constellations were attempted. Multiple 1990s projects did it (Iridium, GlobalStar…) and went bankrupt.
It took 30 years to make the concept work. SpaceX investors seem to be assuming the space data center business will be immediately viable.
Based on very specific assumptions: “…the world in which AI demand is so overwhelming as to exceed the already formidable datacenter capacity additions” — but also this same world is one where GPU chip supply is abundant, there just isn’t enough data centers to put them in.
This does not seem like the likeliest scenario to me.
Agreed, how could we not have datacenter capacity for the GPUs when Meta has shown that you can go from a bare field to an operational datacenter in about 3 months by using tents instead of buildings?
How hard would it be to weatherproof a GPU computing rack? Like how much more cost would it add? So theoretically you could even forgo tents. Just have them at field. Technically you could even maybe run them in freeports. Thus saving any tariff costs...
it sort of makes sense, if the need for that compute is in space and not terrestrial.
In my mind its the same sort of thing as mining in space. it makes no sense to mine ores in space for delivery to earth (unless its something exotic that you cant get on earth).
Mining in space is best used for manufacturing in space (and furthering building in space)... then the cost benefit ratio suddenly flips hard in the "worth it" direction.
Yeah, there are some established needs for compute in space (edge processing for large datasets collected in space, and autonomous spacecraft control) that already happen to some extent and will happen a lot more with greater volumes of EO/SSA data collected and increasing militarization of space. They just don't look much like a datacentre for inference compute
What I don't understand about building a space data center is that you need radiators to release heat. Otherwise, it will become a space thermos. What's even more incomprehensible is that you would need specialized equipment for space radiation, and GPUs are consumables. To make that profitable, you would need pricing that is many times higher than the cost of a regular data center. I don't understand why there are people who actually fall for this. If I say this, people will call me someone who mocks others' challenges, but it seems like they're saying physical problems can be overcome too easily.
The Outer Space Treaty [1] says "A State Party to the Treaty on whose registry an object launched into outer space is carried shall retain jurisdiction and control over such object", so no escaping jurisdiction.
You don't get that with the current plans which require them to have FCC licences and be constantly replacing them by launching from the United States though...
If that's the case, wouldn't it be better to just put it in the desert? Realistically, if noise from calculations is the problem, placing it in a remote area would be more economical.
This also seems more sensible than space, although I guess if you’re in international waters then no military will protect you and someone could eventually destroy it. Way more organizations have the ability to blow up a boat is compared to a space station.
I still don’t see what the advantage is. Of course it’s physically possible to build a datacenter in space, but I can’t imagine land prices being that high that the same data center on earth wouldn’t be cheaper. Even just due to launch costs and the more sophisticated equipment needed for space.
I doubt it'll make sense any time soon, but some arguments I can think of are that solar in space can easily be ~50% more efficient at any moment while also being continuous (enough) in the right orbit.
An even more radical idea is to put nuclear in space which would sidestep all the earthly hurdles (beyond the launch).
Right, but even if you get 100% solar time in orbit and maybe 20% on the ground, I still don’t see it. Just from a procurement cost and maintenance standpoint. I think spreading a few datacenters around the world to have quasi continuous availability is easier than launching them on satellites.
> An even more radical idea is to put nuclear in space which would sidestep all the earthly hurdles (beyond the launch).
That makes even less sense to me. Why would you launch then and not just stay on the ground? Do you think a country would allow you to launch a rocket with a nuclear reactor from their land but the reactor is so unsafe that you’re not allowed to operate it on the ground?
Then I would just say put it on a boat and park it in international waters, that’s surely cheaper than orbit, right?
One advantage that does come to mind in light of the Iran war (and the loss of an AWS DC) is difficulty in attacking it, even when it’s directly above foreign territory. I wonder if one of the intended customers will be gov/military? Conjoined spy satellite/DC for some function maybe?
The real issue is that the power situation in LEO is still actually terrible! Your solar is a little more performant, but you're plunged into hard shade every 45 minutes.
There exist the so-called Sun-synchronous orbits, which exploit the precession effect caused by the fact that the Earth is not a sphere, to pass over the same point of the Earth at the same local hour. On a small subset of these Sun-synchronous orbits the Sun is always visible from the satellite (i.e. on the subset of orbits whose plane is approximately perpendicular on the radius that connects the Sun to the Earth).
However, the number of slots that are available in Sun-synchronous orbits with permanent view of the Sun is limited, and many potential users want them. So those who desire to build datacenters would have to compete for such orbital slots. There are much less such slots than for geosynchronous orbits.
Improved control of the satellites for collision avoidance could allow smaller slots, but maneuvering heavy datacenters would require a lot of fuel.
They are mostly planning sun synchronous orbits afaik. That means the orbit is tilted so the earth’s deformed shape continually moves the orbital plane so the satellite is always in the sun (or generally the plane has the same angle to the sun).
I think calling solar a little more performant is underselling it. Once you have LEO getting to a better orbit costs relatively little. Getting from LEO to the moon is only like 30% more than getting from ground to LEO.
Whyyyy are we not building distributed data centers under driveways? I want one under my driveway to melt the snow. It can use my power and water hookups if it pays for them.
Don't we also have to worry about heating of the solar panels themselves? 150W/m^2 isn't the incident power, it's the output power. Incident is something like ten times that. Some of that's going to be reflected, but not all of it.
If I had a dollar for everytime someone told me they have an advanced physics degree and can unequivocally gaurantee me that it will never be physically feasible to cool GPU's in space.... How soon those people have disappeared!
Seems reasonable that the area needed would be less than the solar panels. Since it sould be more efficient to dump heat than collect energy from light.
It seems to me that there’s a lot of math in the middle of this that just doesn’t matter that much: every photon hitting every part of the array is going to add energy. Whatever percent we transform into data doesn’t matter — all the energy from all those photons minus energy used for station keeping needs to be radiated away. I guess if some is used for station keeping so much the better.
I was hoping to read about what exactly these ‘heat pump radiators’ look like, but I guess ultimately they’re going to be lasers or flashlights or some such thing.
A relativity course question I recall from my youth asked how long an astronaut stranded 1km from a ship would need to point the flashlight away to return to the ship based on the mass of the photons leaving the light — spoiler - this can work in time to save an astronaut’s life — double spoiler: if you’re really accurate, it’s better to throw the flashlight if you’re in a hurry.
As I write this I realize my physics model is super weak, because I’m not sure what percent of the energy used to make a photon turns into the photon’s mass (and therefore is pushing against the laser), and what is in light and therefore just, you know, carries on for billions of years until it hits something else.
This isn't terribly practical. Yes, we can deal with heat. The trouble is cost, and dealing with high energy radiation both flipping bits and corrupting the silicon.
I think it's a vary valid option to launch swarms of datacenters into space. I think a few decades to a hundred years from now, it will be the norm. Until then, we can find plany of land to do it. Instead a launch, you just need a battery. Much cheaper. All the rest stays the same.
All the comments are negative so I'll play the devil's advocate. Here's the steelman case for SpaceX orbital datacenters.
# EARTH IS FULL
It sounds ridiculous but the ability to build AI datacenters on Earth is nearly exhausted. The options are:
• USA, Australia. The electricity infrastructure has already bottlenecked and some datacenters like Colossus are being forced to build their own power plants, but that's also bottlenecked on gas turbine capacity. Hacks like recycling jet turbines only squeeze a bit more out. Terrestrial solar can't be used to escape this problem because you need the clusters to run at night too.
• Europe. Deindustrialized, EU Commission is anti AI, very expensive power, grid also bottlenecked. Forget it.
• Middle East. Had some datacenters until they got blown up by Iran.
• China. Got power but bottlenecked by trade sanctions. Might well do a big buildout when Ascend starts to be competitive, but Chinese demand is likely to absorb it.
• Latin America, Africa, south-east Asia, etc: bottlenecked by political stability, not pro-business enough, etc.
In space you don't need gas turbines because solar can be 24/7, political risks aren't there, you aren't bottlenecked by grid capacity. Even if it costs more to put stuff in space that doesn't matter if space is the only place you can put stuff.
# INFERENCE NOT TRAINING
Trying to do backpropagation in space would be a bad idea. You need extreme locality in a single physical location for networking reasons. But a lot of modern AI load including training load is just inference, which only requires small pods not entire clusters, and bandwidth needs in/out aren't that high. Inferencing can fit on a satellite.
# COOLING
As the article lays out this isn't necessarily the problem people are assuming. Also there are candidate designs from decades ago for ferrofluid droplet radiators. These might be overkill but can in theory radiate huge amounts of heat without needing to launch big radiators.
# COST
Unlike terrestrial data centers which are always bespoke projects, inferencing satellites can be mass manufactured. SpaceX and Elon in general are good at setting up mass production lines, and it seems apparent that SpaceX has no intention of throwing very high margin Nvidia hardware into orbit. The plan is to use Tesla's AI chips i.e. SpaceX could acquire accelerators at cost. This changes the cost calculations quite a bit. Although these accelerators might not be useful for training or research, most training workloads would stay on Earth so that doesn't matter (the inferencing loads moved into space would free up terrestrial hardware for training anyway).
The real wild card is if there's enough demand for a 'good enough' model that it's predicted to last the lifetime of the satellite. In that case the weights could be fabbed directly into the chips like Taalas does, and so the energy consumption would be far lower.
# BUSINESS CASE
It's possible that datacenter construction goes the same way as nuclear and becomes impossibly expensive here on Earth. If so then SpaceX can end up with a near monopoly on new inferencing capacity, making them the gateway to AI and the new Nvidia.
What's especially confounding is that the mere existence of orbital inferencing might actually create that outcome, because politicians would find it much easier to squash datacenter / power projects to please activists if there is a genuine alternative!
It’s an interesting era to be alive in, where the rich and powerful (Trump, Musk) just assert things evidence free and then a whole community rolls in to come up with a post-facto rationalisation.
Musk hasn’t thought about any of this. He just says stuff. He agreed with a statement that “space cooling is free”[1], as a sign of how deeply considered it all is.
Only stating the solar panel - radiator area ratio and not mentioning the actual area that a data center should have, is a sin of omission - and a great one. Also, hand-waving numerous engineering problems just by saying “we’ve solve harder problems before / you can come up with a few designs” is the last thing I want to see in anything that comes up in Hacker News.
Allow me to propose a modest alternative to space data centers, namely mountaintop data centers. This would consist of a container full of servers and GPUs and what else goes into a data center, a wind turbine for power and a communication module (say laser or microwave) for communicating with a base station with a fiber connection. This would be lifted on top of a mountain by a helicopter and bolted in place. Cooling would be provided by heat sinks exposed to the outside air. Some of the nodes could relay traffic from other nodes on remote mountain tops out of sight of the base station.
This scheme has many advantages over space data centers including launch costs, cooling, connection latency, servicability and ease of recycling.
I think we can't rule out the explanation that all the ideas of space data centers could be connected to a desire by some of finding additional applications for rockets that can transport stuff to space.
I think we can't rule out the explanation that all the ideas of space data centers could have been connected solely to a desire to pump SpaceX's IPO.
It also provides a post hoc rationale for rolling Elon's loss making businesses into SpaceX. The bull case for ODCs looks a lot like the bull case for space based solar power that Musk once called "the stupidest thing ever"...
That said, SpaceX aren't the only entity proposing ODCs, they're just the only ones promising they're going to make country-sized profits out of them...
A better idea is to put them on Mars, so people actually think before they send a stupid question.
I'm thinking if we send them out to catch the Voyager probe up, people might have time to write stuff themselves before they ask the computer to do it for them :)
What about the rough weather and difficulty of maintenance, especially in rough weather?
Maintenance for a mountaintop data center only requires a team of skilled mountaineers. In space you'd need astronauts. It's at least an order of magnitude cheaper, perhaps two or three.
What advantage does a mountaintop have versus a more accessible earth-based location?
I suspect one of the motivations for space datacenters is to try to stay out of the reach of all jurisdictions so you Musk can start to run his companies as an autonomous state.
If the goal is to avoid various national jurisdictions, what about floating a barge out into international waters? Do it near the equator in the middle of the pacific to maximize the difficulty of humans getting there. Use the money saved by not needing to launch into orbit to purchase massive gun turrets to prevent piracy.
What would be accomplished by doing this vs placing them basically anywhere else on earth?
It is a comment on the absurdity of orbital data centers. Mountaintop data centers sound absurd, but are more feasible and efficient than orbital ones in nearly every aspect.
https://en.wikipedia.org/wiki/A_Modest_Proposal
Space data centers are physically possible but don't make financial sense. The total cost of an orbital datacenter over five years is at least 2–3x that of a terrestrial one.
But those economics don't matter to SpaceX, because the main purpose of its orbital data centers is to create a use case for Starship. Starship has to fly frequently to iron out the kinks, encounter and fix rare (1/1000) failure situations, and optimize the launch cadence which pushes launch costs down. Plus Starship needs to fly a lot before it's ready for crewed flight. The long-term goal is a Starship optimized for crewed interplanetary travel. Orbital data centers are a payload that bring in some revenue, and provide a reason to launch constantly.
It's the same thing they did with Starlink to make Falcon 9 as reliable and rapidly reusable as it is.
2-3x sounds like a very low estimate.
There’s so much data center capacity being built all over the Earth. Thousands of large projects across US / China / Europe / Middle East. It would be astonishing if something that’s never been done before could be so cost-competitive immediately.
Starlink wasn’t the first time LEO communications constellations were attempted. Multiple 1990s projects did it (Iridium, GlobalStar…) and went bankrupt.
It took 30 years to make the concept work. SpaceX investors seem to be assuming the space data center business will be immediately viable.
SemiAnalysis' report on orbital data centers estimated 4x terrestrial costs in 2026, then parity around ~2040.
https://newsletter.semianalysis.com/p/to-boldly-go-the-case-...
Based on very specific assumptions: “…the world in which AI demand is so overwhelming as to exceed the already formidable datacenter capacity additions” — but also this same world is one where GPU chip supply is abundant, there just isn’t enough data centers to put them in.
This does not seem like the likeliest scenario to me.
Agreed, how could we not have datacenter capacity for the GPUs when Meta has shown that you can go from a bare field to an operational datacenter in about 3 months by using tents instead of buildings?
How hard would it be to weatherproof a GPU computing rack? Like how much more cost would it add? So theoretically you could even forgo tents. Just have them at field. Technically you could even maybe run them in freeports. Thus saving any tariff costs...
it sort of makes sense, if the need for that compute is in space and not terrestrial.
In my mind its the same sort of thing as mining in space. it makes no sense to mine ores in space for delivery to earth (unless its something exotic that you cant get on earth). Mining in space is best used for manufacturing in space (and furthering building in space)... then the cost benefit ratio suddenly flips hard in the "worth it" direction.
Yeah, there are some established needs for compute in space (edge processing for large datasets collected in space, and autonomous spacecraft control) that already happen to some extent and will happen a lot more with greater volumes of EO/SSA data collected and increasing militarization of space. They just don't look much like a datacentre for inference compute
My guess is: it just sounds cool.
Like the cybertruck
What I don't understand about building a space data center is that you need radiators to release heat. Otherwise, it will become a space thermos. What's even more incomprehensible is that you would need specialized equipment for space radiation, and GPUs are consumables. To make that profitable, you would need pricing that is many times higher than the cost of a regular data center. I don't understand why there are people who actually fall for this. If I say this, people will call me someone who mocks others' challenges, but it seems like they're saying physical problems can be overcome too easily.
I have the feeling that the only reason that might actually be done is to escape from any kind of jurisdiction. In space no one can hear you compute.
The Outer Space Treaty [1] says "A State Party to the Treaty on whose registry an object launched into outer space is carried shall retain jurisdiction and control over such object", so no escaping jurisdiction.
[1] https://en.wikipedia.org/wiki/Outer_Space_Treaty, https://en.wikisource.org/wiki/Outer_Space_Treaty_of_1967#Ar...
You don't get that with the current plans which require them to have FCC licences and be constantly replacing them by launching from the United States though...
If that's the case, wouldn't it be better to just put it in the desert? Realistically, if noise from calculations is the problem, placing it in a remote area would be more economical.
It's a play on "in space no-one can hear you scream", not a literal "compute is so noisy we need to go to space"
The jurisdiction issue is the bigger one. Deserts don't solve this; International waters, possibly do, but then you've got other issues.
Is that so? How embarrassing. I'm not a native English speaker, so I didn't understand the metaphor.
The desert is still under someone’s jurisdiction. Perhaps the best on-planet comparison is creating a man made island in the middle of the ocean.
This also seems more sensible than space, although I guess if you’re in international waters then no military will protect you and someone could eventually destroy it. Way more organizations have the ability to blow up a boat is compared to a space station.
I still don’t see what the advantage is. Of course it’s physically possible to build a datacenter in space, but I can’t imagine land prices being that high that the same data center on earth wouldn’t be cheaper. Even just due to launch costs and the more sophisticated equipment needed for space.
I doubt it'll make sense any time soon, but some arguments I can think of are that solar in space can easily be ~50% more efficient at any moment while also being continuous (enough) in the right orbit.
An even more radical idea is to put nuclear in space which would sidestep all the earthly hurdles (beyond the launch).
Right, but even if you get 100% solar time in orbit and maybe 20% on the ground, I still don’t see it. Just from a procurement cost and maintenance standpoint. I think spreading a few datacenters around the world to have quasi continuous availability is easier than launching them on satellites.
> An even more radical idea is to put nuclear in space which would sidestep all the earthly hurdles (beyond the launch).
That makes even less sense to me. Why would you launch then and not just stay on the ground? Do you think a country would allow you to launch a rocket with a nuclear reactor from their land but the reactor is so unsafe that you’re not allowed to operate it on the ground?
Then I would just say put it on a boat and park it in international waters, that’s surely cheaper than orbit, right?
One advantage that does come to mind in light of the Iran war (and the loss of an AWS DC) is difficulty in attacking it, even when it’s directly above foreign territory. I wonder if one of the intended customers will be gov/military? Conjoined spy satellite/DC for some function maybe?
The real issue is that the power situation in LEO is still actually terrible! Your solar is a little more performant, but you're plunged into hard shade every 45 minutes.
There exist the so-called Sun-synchronous orbits, which exploit the precession effect caused by the fact that the Earth is not a sphere, to pass over the same point of the Earth at the same local hour. On a small subset of these Sun-synchronous orbits the Sun is always visible from the satellite (i.e. on the subset of orbits whose plane is approximately perpendicular on the radius that connects the Sun to the Earth).
However, the number of slots that are available in Sun-synchronous orbits with permanent view of the Sun is limited, and many potential users want them. So those who desire to build datacenters would have to compete for such orbital slots. There are much less such slots than for geosynchronous orbits.
Improved control of the satellites for collision avoidance could allow smaller slots, but maneuvering heavy datacenters would require a lot of fuel.
They are mostly planning sun synchronous orbits afaik. That means the orbit is tilted so the earth’s deformed shape continually moves the orbital plane so the satellite is always in the sun (or generally the plane has the same angle to the sun).
I think calling solar a little more performant is underselling it. Once you have LEO getting to a better orbit costs relatively little. Getting from LEO to the moon is only like 30% more than getting from ground to LEO.
True, but lifting the fuel to power that “small” orbital boost is unintuitively expensive
(Refer to the tyranny of the “Rocket Equation”)
There was also this[1] analysis in late 2025 from a former NASA engineer that looked at the impracticability.
[1] https://taranis.ie/datacenters-in-space-are-a-terrible-horri...
Whyyyy are we not building distributed data centers under driveways? I want one under my driveway to melt the snow. It can use my power and water hookups if it pays for them.
Aren't some bitcoin farms used for heating in some places, like agriculture in greenhouses?
Don't we also have to worry about heating of the solar panels themselves? 150W/m^2 isn't the incident power, it's the output power. Incident is something like ten times that. Some of that's going to be reflected, but not all of it.
If I had a dollar for everytime someone told me they have an advanced physics degree and can unequivocally gaurantee me that it will never be physically feasible to cool GPU's in space.... How soon those people have disappeared!
Seems reasonable that the area needed would be less than the solar panels. Since it sould be more efficient to dump heat than collect energy from light.
It seems to me that there’s a lot of math in the middle of this that just doesn’t matter that much: every photon hitting every part of the array is going to add energy. Whatever percent we transform into data doesn’t matter — all the energy from all those photons minus energy used for station keeping needs to be radiated away. I guess if some is used for station keeping so much the better.
I was hoping to read about what exactly these ‘heat pump radiators’ look like, but I guess ultimately they’re going to be lasers or flashlights or some such thing.
A relativity course question I recall from my youth asked how long an astronaut stranded 1km from a ship would need to point the flashlight away to return to the ship based on the mass of the photons leaving the light — spoiler - this can work in time to save an astronaut’s life — double spoiler: if you’re really accurate, it’s better to throw the flashlight if you’re in a hurry.
As I write this I realize my physics model is super weak, because I’m not sure what percent of the energy used to make a photon turns into the photon’s mass (and therefore is pushing against the laser), and what is in light and therefore just, you know, carries on for billions of years until it hits something else.
This isn't terribly practical. Yes, we can deal with heat. The trouble is cost, and dealing with high energy radiation both flipping bits and corrupting the silicon.
I think it's a vary valid option to launch swarms of datacenters into space. I think a few decades to a hundred years from now, it will be the norm. Until then, we can find plany of land to do it. Instead a launch, you just need a battery. Much cheaper. All the rest stays the same.
All the comments are negative so I'll play the devil's advocate. Here's the steelman case for SpaceX orbital datacenters.
It sounds ridiculous but the ability to build AI datacenters on Earth is nearly exhausted. The options are:• USA, Australia. The electricity infrastructure has already bottlenecked and some datacenters like Colossus are being forced to build their own power plants, but that's also bottlenecked on gas turbine capacity. Hacks like recycling jet turbines only squeeze a bit more out. Terrestrial solar can't be used to escape this problem because you need the clusters to run at night too.
• Europe. Deindustrialized, EU Commission is anti AI, very expensive power, grid also bottlenecked. Forget it.
• Middle East. Had some datacenters until they got blown up by Iran.
• China. Got power but bottlenecked by trade sanctions. Might well do a big buildout when Ascend starts to be competitive, but Chinese demand is likely to absorb it.
• Latin America, Africa, south-east Asia, etc: bottlenecked by political stability, not pro-business enough, etc.
In space you don't need gas turbines because solar can be 24/7, political risks aren't there, you aren't bottlenecked by grid capacity. Even if it costs more to put stuff in space that doesn't matter if space is the only place you can put stuff.
Trying to do backpropagation in space would be a bad idea. You need extreme locality in a single physical location for networking reasons. But a lot of modern AI load including training load is just inference, which only requires small pods not entire clusters, and bandwidth needs in/out aren't that high. Inferencing can fit on a satellite. As the article lays out this isn't necessarily the problem people are assuming. Also there are candidate designs from decades ago for ferrofluid droplet radiators. These might be overkill but can in theory radiate huge amounts of heat without needing to launch big radiators. Unlike terrestrial data centers which are always bespoke projects, inferencing satellites can be mass manufactured. SpaceX and Elon in general are good at setting up mass production lines, and it seems apparent that SpaceX has no intention of throwing very high margin Nvidia hardware into orbit. The plan is to use Tesla's AI chips i.e. SpaceX could acquire accelerators at cost. This changes the cost calculations quite a bit. Although these accelerators might not be useful for training or research, most training workloads would stay on Earth so that doesn't matter (the inferencing loads moved into space would free up terrestrial hardware for training anyway).The real wild card is if there's enough demand for a 'good enough' model that it's predicted to last the lifetime of the satellite. In that case the weights could be fabbed directly into the chips like Taalas does, and so the energy consumption would be far lower.
It's possible that datacenter construction goes the same way as nuclear and becomes impossibly expensive here on Earth. If so then SpaceX can end up with a near monopoly on new inferencing capacity, making them the gateway to AI and the new Nvidia.What's especially confounding is that the mere existence of orbital inferencing might actually create that outcome, because politicians would find it much easier to squash datacenter / power projects to please activists if there is a genuine alternative!
Note: I'm not invested in SpaceX.
It’s an interesting era to be alive in, where the rich and powerful (Trump, Musk) just assert things evidence free and then a whole community rolls in to come up with a post-facto rationalisation.
Musk hasn’t thought about any of this. He just says stuff. He agreed with a statement that “space cooling is free”[1], as a sign of how deeply considered it all is.
[1] https://nitter.net/elonmusk/status/1998483552669937682?lang=...
See also https://andrewmccalip.com/space-datacenters with some models and numbers you can play with.
See also:
https://www.youtube.com/watch?v=FlQYU3m1e80
And also: https://www.youtube.com/watch?v=JAcR7kqOb3o
I'll add another to the list of relevant links in the comments: https://spectrum.ieee.org/orbital-data-centers-heat
[dead]