Cooling in Space(guille.site) |
Cooling in Space(guille.site) |
The incoming power is not the electrical power generated by the solar panels, but the entire power of the light that is absorbed by the solar panels and by the body of the satellite.
Even with a perfectly reflecting body and with SOTA solar panels, the amount of incoming power is at least double in comparison with the electrical power consumed by the datacenter.
Also, the heat radiated is smaller than in TFA, because no radiator is perfectly black at the radio waves in the frequency range corresponding to the ambient temperature.
I am too lazy to make the correct computation, but there was another article linked on HN some days ago where a more plausible computation was done and the conclusion was that the minimum area of the radiators is slightly larger than the area required for solar panels.
This would still be feasible, but in reality the area would have to be even larger, because the radiator cannot have a uniform temperature, the parts where the cooling fluid is incoming will be hotter than the parts where the cooling fluid is outgoing. Moreover, the pumping of the cooling fluid requires extra power that must be added to the power budget.
There is no doubt that it is possible to build a space datacenter, if much more GPUs are installed in it than necessary, to enable to correct the more severe transient errors and to preserve enough capacity after many GPUs become permanently defective, but the cost will not be competitive with terrestrial datacenters any time soon.
This scheme has many advantages over space data centers including launch costs, cooling, connection latency, servicability and ease of recycling.
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.
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.
https://newsletter.semianalysis.com/p/to-boldly-go-the-case-...
Like the cybertruck
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.
[1] https://en.wikipedia.org/wiki/Outer_Space_Treaty, https://en.wikisource.org/wiki/Outer_Space_Treaty_of_1967#Ar...
[1] https://taranis.ie/datacenters-in-space-are-a-terrible-horri...
The one ESA used to use is called
ESATAN
I used to share office with a colleague who mostly worked with it. This was at a time when the fad of naming products with an initial "e" had just faded. The surviving victim is "ebay" I guess, but at a time there were many like it.
So my natural reading of the huge ASCII art rending on ESATAN's title screen has always been E-SATAN. Sorry ESA.
Joking aside, you can download ESATAN here:
It's not obvious what the download allows but expect restrictions. I remember the version we used had a HW dongle with heavy price tag every month.
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.
You can bullshit humans but you can't bullshit reality.
"Cooling is actually much easier in space than it is on earth. You can just radiate to the vacuum."
I don't think that follows. The radiator is only the final heat sink. You still need to move heat from very dense chips into a deployable, space-rated radiator, and handle pumps, loops, leaks, redundancy, radiation damage, replacement, eclipses, Earth IR/albedo, and launch mass.
AI sat mini can use a simpler single ammonia loop, since the ISS uses a water loop on the station side to avoid toxicity issues in case of a coolant leak.
It's a far simpler engineering problem to solve compared to other challenges SpaceX is facing (Starship, Raptor, Starlink).
# 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.Space radiation isn't necessarily a problem. Bit flips can be tolerate to quite a high degree for inferencing because models can recover from corruptions in the activation stream or even some bad tokens.
# 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!
Note: I'm not invested in SpaceX.
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=...
An even more radical idea is to put nuclear in space which would sidestep all the earthly hurdles (beyond the launch).
> 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?
I see the point to maybe do some onboard data processing on spy satellites etc, but on the other hand, downlink bandwidth seems to be become less of an issue over time, so it doesn’t seem that important to me compared to just sending down the raw data over star link or the military equivalent.
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. Other countries would certainly be outraged if USA occupied all the available slots with datacenters.
Improved control of the satellites for collision avoidance could allow smaller slots, but maneuvering heavy datacenters would require a lot of fuel, so they might require periodic refueling, greatly increasing the costs.
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...
(Refer to the tyranny of the “Rocket Equation”)
But space based dc accomplish something that mountaintop dc do not. The different list of benefits/tradeoffs are why space DC are proposed and mountaintop ones are not. It's a difference of kind, not degree. It's not a meaningful experiment to just try to build DC in hard places and then we can finally validate space.
Stated benefits in particular:
- Power available 24/7 for "free"
- coms w/o interruption using existing infra
- Rideshare (SPX can build out capacity while other lifts pay some of the bill for lift)
- Nonregulation
- Very low latency to "places of interest far from USA mountains"
And no, I do not believe that mountaintop automatically satisfies these benefits in a smooth way such that mountaintop is a meaningful stepping stone towards space.
The Sun is visible from Earth as well, the last time I checked.
In LEO you don't get power 24/7 because you are only 500km above the Earth. Yes the Sun is more attenuated on Earth but what we care about is $/W not raw wattage, and Earth certainly has cheaper $/W than space.
> - coms w/o interruption using existing infra
I'm perplexed how comms might be easier in space than on Earth where you can just run a cable.
> - Rideshare (SPX can build out capacity while other lifts pay some of the bill for lift)
On Earth you don't need to rideshare because you don't have to ride a rocket.
> - Nonregulation
Space is more regulated than Earth. The only way to get to space is via a rocket which is the same as an ICBM. Governments regulate the process of building ICBMs and what payloads can ride on them.
If you want non-regulation then go to international waters or find a bribable government.
> - Very low latency to "places of interest far from USA mountains"
The latency is not terrible in LEO but it's nowhere near as good as on Earth.
But it's also irrelevant: all your infrastructure supporting such a thing, including your ability to fund it, is on Earth, in someone's jurisdiction.
The US government is hardly going to say "well the datacenter is in space, guess there's nothing we can do about the owner who lives in California..."
Nuclear fission reactors, similar to those used on submarines or ships, would enable very different applications.
Until now, they have not been used for fear of what would happen after a failed rocket launch, when the reactor would fall back on Earth.
This could be mitigated by sending only components of the reactor and assembling it in space.
I do not think that routine exploration of the Solar System beyond Mars will ever be possible without using at least nuclear fission reactors, because it is too slow with chemical sources of energy.
https://en.wikipedia.org/wiki/TOPAZ_nuclear_reactor#TOPAZ-I
> This could be mitigated by sending only components of the reactor and assembling it in space.
How would that help? The main problem if the launch fails is that the radioactive material will spread around (in the worst case, they are encased so exactly that doesn’t happen).
You could maybe spread out the nuclear material in multiple launches but you would just increase the risk of a small contamination in exchange for a smaller risk of a large contamination.
I don’t think the individual parts of the reactor are intrinsically safer than the reactor, it’s not like it’s going to become an atomic explosion during launch.
And here's the thing: all of this is competing with solar+batteries cost on Earth. The power situation is the only advantage here.
Like why not put a datacenter on a barge and run an HVDC line out to it far offshore? That would be expensive...but more expensive then space? It's not even outside of the capability set of SpaceX, who already run drone ships to facilitate Falcon 9 landings.
No it’s not, it’s not either SSO or LEO. You can have SSO at 600km which is lower than the normal LEO satellite.
I agree that it doesn’t make a lot of sense (look at the root comment of this thread), but you absolutely can make a LEO datacenter with 100% sun coverage if you want to.
(Hint to the downvoters: I'm not being serious.)
blub °Oo.
And absolutely no one, anywhere, ever, has the capability to damage or destroy a satellite...
Not necessarily in orbits which are easily reachable for current ASATs, nor 'economical'.
For now...
If you could pay a few space sherpas $100k to head up into LEO and service the thing, it would definitely be worth it.
> If you could pay a few space sherpas $100k to head up into LEO and service the thing, it would definitely be worth it.
Would it? Whatever you pay to launch the repair tech plus the replacement parts could instead be spent launching new hardware. Obviously the repair payload is a fraction of the total weight of new hardware but is it a small enough fraction to make repairing things worthwhile? I think it's likely that disposable is cheaper in this scenario.
This does not seem like the likeliest scenario to me.
The jurisdiction issue is the bigger one. Deserts don't solve this; International waters, possibly do, but then you've got other issues.
really think about that statement when discussing deliberately avoiding government jurisdiction...
(perhaps also consider that it is not the case that no one can damage a lot of satellites in orbit, but that up until recently no one has had any incentive to build the number of interceptors you would need to do it. But how viable is a space-based datacenter business if you decide to try and pretend you're untouchable, and one of the _many_ governments which operates anti-satellite weapons simply shoots one of your satellites? The debris field from ASAT weapons tests has been of considerable concern everytime they've been used - and given the proximity of useful orbital slots for such a service, the number of intercepts required to render a constellation completely inoperable is going to be _far less_ then the number of satellites).
(in the vein of motivation too: it is well within the power of most well-funded governments to build laser systems would would degrade or destroy orbital satellites, but again, no one has had considerable motivation to do so till recently)
(and of course all of this is - again - competing against the simpler option of simply arresting the people on Earth, or interdicting their ground stations)
The main problem here is that it reduces efficiency (cooling a large datacenter is more efficient per Watt of dissipated heat than a shipping container) and increases initial cost (building in a shipping container is not actually as easy as doing it in a normal-ish building).
Portability (when offline, you can put a shipping container like this on a truck and cart it around) and availability (no need for a new/refit building, only power is required and could be included in the container with a generator (gas/diesel)) are the main reasons for accepting a higher TCO here.