Thread

Google’s Project Suncatcher to put ML infrastructure in space

233 points · 553 comments · xnx

zactato · 1790267254 · focus · HN ↗
How are they solving the heat dissipation issues?
1. GMoromisato · 1790270501 · focus · HN ↗
  Please stop with this argument. This is one of the easiest problems to solve. Heat dissipation requires a radiator, which is just a dumb hunk of metal. All you need is a cheap way to launch mass to orbit, which is exactly what SpaceX (and Chinese companies) are doing.
  The only valid argument against data centers in space is the economic cost. If the per megatoken price for space datacenters is higher than terrestrial, then this won't work.
  But the price for terrestrial datacenters keeps going up and technology keeps dropping the price of space-based.
  1. 0cf8612b2e1e · 1790272676 · focus · HN ↗
    That is not a fake argument, but real physics. Yes, you can design out X kw can be dissipated by this much radiators, but that adds an enormous quantity of mass, more than the solar panels that feed it.
    If you are trying to generate a profit, every extra kg puts you further in the red. It still costs over $1000/kg of mass to get into orbit.
    GMoromisato · 1790275041 · focus · HN ↗
    So is it a physics problem or an economic problem? Sounds to me like you are acknowledging that it is just an economic problem. If it cost $1/kg to get to orbit then this wouldn't be a problem, right? Just make a bigger radiator.
    But if it's an economic argument, then you need to do the actual math. How big is the radiator? How low could the price to orbit go? How much can you charge per million tokens in 2030?
    The reason this is a fake argument is because the validity depends on the math, and nobody advancing the "you can't cool stuff in space" argument is actually doing the math.
    motionlessveloc · 1790281005 · focus · HN ↗
    The economics is constrained by physics. AI in space is not viable if it costs $1000/kg to launch to space. Starship promises to cut that down to $100/kg (more if you believe Elon, but most don't), but that's still not competitive.
    Sure, things would be different if the cost was $1/kg, but short of somebody building a space elevator that's just not going to happen.
    GMoromisato · 1790282269 · focus · HN ↗
    I won't believe less than $100/kg until I see it. I agree with you on that.
    But are you sure that $100/kg is not competitive? I don't remember all the math, but even their initial AI1 design[1] would throw a lot of profit, if you can sell at the price they offered to Anthropic.
    I suspect the price of manufacturing the satellite, plus chips, is the dominating factor, not necessarily the launch costs.
    [1] <a href="https://www.spacex.com/spacexai/starmind" rel="nofollow">https://www.spacex.com/spacexai/starmind
    sourcecodeplz · 1790320032 · focus · HN ↗
    > if you believe Elon.
    Why wouldn;t you? he delivered on almost everything he promised, just the timeline was a lot later.
    we are talking sci-fi things here, these will take time but can be done.
    would you rather we bury our heads in the sand and never innovated beyond basics?
    cyberax · 1790282091 · focus · HN ↗
    It's a fundamental physics problem. You need to have huge radiating surfaces.
    A 1GW datacenter with chips operating at 100C (which is probably doable) will need a radiator that has a surface of one square _kilometer_, and this is with all the favorable assumptions. Realistically you'll need about 2x of that.
    If you want your DCs to be on a 1000km orbit (for reasonable ping times), you'll be able to _resolve_ these satellites with a naked eye!
    Sorry. But this idea is fundamentally unworkable.
    GMoromisato · 1790282643 · focus · HN ↗
    The current plan is for 250 kW peak with 160 m^2 radiators. It will weigh about 4 tons so you can pack 25 on one flight of Starship.
    Deploy 4,000 and you're at 1 GW. That's 160 launches.
    BTW: SpaceX has already manufactured and launched 10,000 Starlink satellites and Falcon 9 launches about 150 times per year. None of this seems unworkable.
    cyberax · 1790285003 · focus · HN ↗
    Can we solve the fertilizer price problem by hauling it with airplanes? Yes, we can! It's easy, just load the potash fertilizer into an airplane and unload it directly into the traincars. I even designed a neat conveyor belt system to speed up unloading!
    Look at the numbers. 1kW of nuclear power capacity on Earth is around $2000, and that's 24/7 guaranteed power. So a 250kW cluster needs $500000 to cover its power demand with near 100% reliability and with some ongoing cost.
    And if we're OK with some interruptions, then we can use solar+wind at around $100000 and with essentially no ongoing cost. If we assume the absolutely best projected launch cost of $100 per kg (vs the current one of ~$800), that's just 1 ton of material in space!
    So you're off by 2-3 orders of magnitude in cost. And this kind of "it's unprofitable" is actually a fundamental issue.
    GMoromisato · 1790293813 · focus · HN ↗
    Okay, so you admit it is possible. Cooling is not the problem. We're making progress.
    Now your argument is that it's unprofitable. Here's the calculator: <a href="https://andrewmccalip.com/space-datacenters" rel="nofollow">https://andrewmccalip.com/space-datacenters
    Specifically, if SpaceX can get the price of a satellite down to $8 per watt (about $2 million USD) then it will compete with terrestrial.
    I just don't understand how you can be so certain that they can't do that. I'm not certain that they can, but being certain that it's impossible seems completely evidence-free.
    cyberax · 1790294863 · focus · HN ↗
    This calcualtor is bullshit (inflated terrestrial costs and underflated orbital costs). It doesn't pass the basic sniff test: $15B for 1GW of terrestrial power is more than enough to build AN ENTIRE 3GWe NUCLEAR POWER PLANT. From scratch. With 75 years of expected life.
    So no, the calculations show that space is NOT feasible unless you want to do that for nefarious reasons: evading regulations, using AI for criminal enterprises, military use, that sort of thing.
    Only these applications have the profit margin that even comes close to justifying it.
    sobellian · 1790289584 · focus · HN ↗
    Okay any argument about why space is uniquely challenging is going to revolve around physics. Sure it's not literally physically impossible, but we need to explain to people why this is different from shipping the GPUs to Ohio.
    If you want math then <a href="https://andrewmccalip.com/space-datacenters" rel="nofollow">https://andrewmccalip.com/space-datacenters exists. The numbers are grim for orbital DC. Even if you drag the launch cost slider all the way to $1/kg (by the way this is literally sci-fi, per ChatGPT air freight of semiconductors from Taiwan to Ohio costs $9/kg and ocean/train freight costs a bit under $1/kg for a reasonable shipment so good luck with $1/kg to LEO this century) it is still more than twice as expensive as terrestrial DCs.
    GMoromisato · 1790293404 · focus · HN ↗
    But that calculator shows most of the cost is in the satellite. At $8 per watt (or about $2 million per satellite) the cost of orbital compute matches terrestrial.
    That sounds absolutely possible. But in any event, we're now arguing a different thing.
    The only thing I'm arguing is that cooling is a solved problem. I don't know if SpaceX will ever get the price down so that it is economical. But I'm convinced that there is no universe in which they hit their foreheads and say, "Oh no, we forgot about cooling!"
    sobellian · 1790295771 · focus · HN ↗
    Note that this calculator is actually quite optimistic for orbital wrt. many things including cooling and effect on launch, as:
    > No additional mass for liquid cooling loop infrastructure; likely needed but not included
    > Thermal: only solar array area used as radiator; no dedicated radiator mass assumed
    In hardware and mfg. solvable vs. solved is a big difference. And I too believe that SpaceX's engineers know about radiator panels. But the more cynical interpretation is that whatever the SpaceX engineers think about the technical merits, they are not being asked for that. They are just being asked for a pretext that justifies the xAI acquisition. Elon is also discussing lunar satellite factories that launch the satellites via railgun. Now, is this physically impossible? No, that isn't physically impossible either and I will seriously defend the physical possibility of this. It's not going to happen though.
    And you could spend all the engineering costs on building some seriously efficient terrestrial DCs, but somehow all these analyses start with "assume that launch and satellite technology advances manyfold and terrestrial DCs stagnate or become less efficient, then if you squint the two numbers get kinda close."
    Ajedi32 · 1790344932 · focus · HN ↗
    Certainly launch has to advance a lot for this to make sense. SpaceX is betting big on Starship getting close to $100/kg. But does satellite tech have to advance at all to make this work? SpaceX has already successfully deployed one megaconstellation into orbit and it's extremely profitable. Isn't this just another Starlink with bigger solar panels and radiators?
    sobellian · 1790347442 · focus · HN ↗
    Yes it has to advance a lot. In a DC all your supporting equipment has returns to scale and can be repaired if it breaks. If a GPU breaks, a sysadmin walks over to the offending rack and swaps the card. In space all that equipment serves just a few cards (this is more like orbital server racks) and it has to work in space (so instead of using an ~infinite heat sink like the Earth, you use radiators etc). We still don't exactly know what effect radiation in LEO would have on stock GPUs - IIRC the experiments to determine this started after Elon went all in on orbital DCs. If anything breaks, you have a flying brick.
    This is why that calculator, even under extremely optimistic assumptions for orbital, and even if you assume launch is zero, still cannot make it competitive with terrestrial.
    Ajedi32 · 1790354768 · focus · HN ↗
    Doesn't Starlink already handle all that though, just with networking racks instead of GPU racks? What's so unique about GPUs compared to other computer equipment, besides the power budget?
    sobellian · 1790356251 · focus · HN ↗
    Commercial GPUs already fail a lot and that's before you put them in LEO. Starlink is doing something that can't be accomplished anywhere else than space. If you could somehow have ground stations that can give wireless internet to every corner of the Earth you would much rather do that. Again, this is all physically possible, but the question is whether it's wise, whether it's saving money or if it's just a giant Rube Goldberg machine. If you assume Starlink satellite $/W then even if launch is zero, it costs twice as much as terrestrial using assumptions charitable to orbital.
    Ajedi32 · 1790357018 · focus · HN ↗
    Fair point about the failure rates. I wonder what causes that, given that there are no moving parts... if it's temperature fluctuations, maybe you could just run the GPU at 100% all the time to avoid that, given that the marginal cost of power for a satellite already in space is $0?
    > If you assume Starlink satellite $/W
    Why would you assume that, given that high power output is not a design goal of Starlink?
    sobellian · 1790357506 · focus · HN ↗
    The calculator I've cited elsewhere uses <a href="https://chatgpt.com/share/69391474-4b24-8005-bb93-ebd4340c65d8" rel="nofollow">https://chatgpt.com/share/69391474-4b24-8005-bb93-ebd4340c65.... The various starlink versions have approximately the same cost per watt across very different power levels. Much of the cost is going to be ~proportional to power - solar panels, heat rejection, power electronics, etc. Listen, if they could make it cheaper to deliver the same amount of power they would. Starlink is approximately all of SpaceX's business.
    Ajedi32 · 1790358636 · focus · HN ↗
    ChatGPT decided to use a constant $800/kg to estimate cost, and you're using that as evidence that cost/W doesn't decrease with scale?
    Anyway, my point is that Starlink is not primarily designed to harvest as much solar power as possible the way data center sats are. I'm sure they're not making it inefficient on purpose, but what you're suggesting is like using the cost/W of a solar powered traffic camera to estimate the cost/W of a solar farm.
    sobellian · 1790359507 · focus · HN ↗
    It's reasonable to 0th order to assume a ball of heat management and phased array antennae has a roughly constant cost per kg, yes. And as I said starlink absolutely wants to use as much power as they can, as communications from space are heavily limited by power.
    Ajedi32 · 1790359958 · focus · HN ↗
    FWIW, SpaceX is estimating their AI sats will have over double the W/kg of what that ChatGPT summary is estimating for Starlink. <a href="https://www.spacex.com/spacexai/starmind" rel="nofollow">https://www.spacex.com/spacexai/starmind Cost is not disclosed.
    sobellian · 1790361698 · focus · HN ↗
    Okay so even assuming all of SpaceX's numbers, which is extremely optimistic, even if we're an angel's advocate and assume that they spend all this money on engineering and optimize their launch costs and their satellite costs and they don't hit any hiccups, and DCs on Earth stay pretty much the same and don't get any added efficiencies. Even then over every time period orbital is still more expensive. And it gets particularly punishing the longer you extend the analysis period as the satellites need to get replaced while the DCs don't spontaneously combust all their infra.
    Ajedi32 · 1790609280 · focus · HN ↗
    I don't think those numbers are optimistic, I suspect they modeled it in CAD and that's what the calculation spat out; it's probably quite accurate.
    Cost is a different matter of course, but it seems incongruous to me that after being confronted with the fact that your power density assumptions were off by a factor of 2x you would continue to insist ChatGPT's weird assumptions about costs scaling linearly with mass are accurate. As was already pointed up up thread, even if all your calculator's assumptions are correct they can still turn a profit if they get the per satellite cost below $2 million.
    sobellian · 1790837162 · focus · HN ↗
    Upthread comment and therefore the premise of your $8/W target is wrong. As I explained, I plugged in all the announced engineering parameters for Starmind, $8/W, and $76/kg. It ties terrestrial over a 5yr period (the slider default) but gets crushed over a 10yr period. Naturally a longer time window is prudent absent some world-ending calamity in 5 years that renders the need for compute irrelevant. And even the 5yr tie is very sensitive to these parameters and requires everything to go right.
  2. mzajc · 1790272720 · focus · HN ↗
    > This is one of the easiest problems to solve. /../ All you need is a cheap way to launch mass to orbit.
    So... not easy? I don't suppose we're at a point where you could reasonably send a large-enough radiator for a multi-gigawatt cluster into space?
    m4rtink · 1790273857 · focus · HN ↗
    Not only that - it would be totally insane to launch something heavy & at the same time fragile from earth (under a lot of vibrations & heavy g-loading).
    This is all a weird speedrun or race. If something we should be working on setting up resource mining from the Moon & asteroids, materials processing on the Lunar surface & in orbit, simple manufacturing in space, etc.
    Instead some people think we can jump straight to a computronium Dyson swarm. :P
    GMoromisato · 1790275481 · focus · HN ↗
    The plan is to launch thousands of 250 kW satellite. Doesn't sound insane at all.
    GMoromisato · 1790275234 · focus · HN ↗
    Now you've moved the goal posts. It's no longer "you can't cool stuff in space"; now it's "we can't launch a 10-gW compute cluster by the end of the year."
    You don't need to send a multi-gigawatt satellite to space. You just need to launch a few thousand 250 kW satellites. That's not against the laws of physics.
  3. teaearlgraycold · 1790272921 · focus · HN ↗
    It will never be cheaper to put compute into orbit. And costs for AI are dropping like a rock here on Earth.
    GMoromisato · 1790277907 · focus · HN ↗
    Never is a long time and you're relying on a bunch of unknowns like the cost of launch to orbit in 2030 and the future regulatory environment here on earth.
    If you know all that out to 2040 then you must be a time traveler. Please try to fix our timeline rather than wasting time on HN.
    teaearlgraycold · 1790278962 · focus · HN ↗
    You’re right. Why am I wasting time on here?
    GMoromisato · 1790295359 · focus · HN ↗
    You're smarter than me on that point. I got sucked into the argument and ended up wasting a couple of hours. I'm an idiot.
    sourcecodeplz · 1790320860 · focus · HN ↗
    no you didnt, i read most of your comments in this thread. very valuable even if little optimistic.
    >> The current plan is for 250 kW peak with 160 m^2 radiators. It will weigh about 4 tons so you can pack 25 on one flight of Starship.
    >> Deploy 4,000 and you're at 1 GW. That's 160 launches.
  4. devmor · 1790273034 · focus · HN ↗
    > Please stop with this argument. This is one of the easiest problems to solve. Heat dissipation requires a radiator, which is just a dumb hunk of metal.
    Convective radiation does not happen in space and this challenge is far more significant than your comment implies. Rather than "a dumb hunk of metal", radiators for spacecraft are often made of ceramics and carbon laminates with higher IR emissivity than convective radiators made of simple metals.
    From the article you're commenting on:
    > The chips can operate for about 15 minutes in space before needing to be shut down so they can cool off, said Travis Beals, Google’s senior director of product management for Project Suncatcher.
    The Thermal Control section on NASA's Small Spacecraft documentation center is quite an interesting read for this subject: <a href="https://www.nasa.gov/smallsat-institute/sst-soa/thermal-control/" rel="nofollow">https://www.nasa.gov/smallsat-institute/sst-soa/thermal-cont...
    ML Infrastructure comes with some pros (larger emissive footprint) and cons (exponentially larger TDP) compared to the concerns there, but if you aren't familiar with the challenges of heat dissipation in space, please give it a read.
    There's also a pretty interesting pop-sci article on cooling the Webb telescope, since it needs to be especially cold for its purpose. Not directly related, but may give insight into both challenges and solutions as well as well. <a href="https://science.nasa.gov/mission/webb/science-overview/science-explainers/how-does-webb-stay-cold/" rel="nofollow">https://science.nasa.gov/mission/webb/science-overview/scien...
    GMoromisato · 1790277636 · focus · HN ↗
    Emissivity is one factor, but it is dwarfed by the T^4 term. Sure, maybe if you use exotic materials you can get from 0.9 to 0.95 emissivity, but why bother? Just run the radiators a little hotter.
    The equation is:
    A ~ (1000 P) / (2 e k T^4)
    Where
    A is the radiator area in square meters P is the power in kilowatts e is emissivity (usually 0.9) k is the constant 5.67e-8
    P and T are the dominating factors. Don't worry about emissivity.
    devmor · 1790278519 · focus · HN ↗
    Emissivity is an important factor here because as I said, and as the sources I linked for you to reference clearly stated convective radiation is not taking place in space.
    Both Google and NASA are worried about this for a reason, if you think they are wrong, you should offer your assistance to them, rather than debating me.
    GMoromisato · 1790279092 · focus · HN ↗
    My point is you don't need fancy/expensive materials for your radiator. Anodized aluminum is at 0.8. With inexpensive coatings you can get to 0.9.
    If your argument against space datacenters is "radiator materials are too expensive" then I just think that's not a very good argument.
    devmor · 1790290906 · focus · HN ↗
    Oh, sure yeah I agree with that. I was just making a point about the “big hunk of metal” comment and the fact that emissivity is important because the expense of radiator materials aren’t the problem - the surface area (and mass) of the radiator is.
    GMoromisato · 1790293966 · focus · HN ↗
    And I agree with you on that. "Big hunk of metal" was too much of an exaggeration--SpaceX's design has liquid cooling, so I assume they have some channels or tubs running through the radiator, plus pumps, etc.
  5. Avicebron · 1790273826 · focus · HN ↗
    Is Kessler syndrome priced into cost? Or is that just like, someone else's problem?
  6. legohead · 1790274545 · focus · HN ↗
    and yet notice how the cooling video / section was the only one they didn't have a solution for... just saying "radiator" doesn't make sense - the radiator heats up too. it's how you get rid of the heat, not where you put it.
    GMoromisato · 1790277423 · focus · HN ↗
    Huh? Radiators are known technology. They have them on ISS; they have them on every Starlink satellite. This isn't like warp drive or antigravity.
    Moreover, you can easily calculate how big of a radiator you need for a given power level and temperature. You can use the Stefan–Boltzmann law:
    A ~ (1000 P) / (2 e k T^4)
    Where
    A is the radiator area in square meters P is the power in kilowatts e is emissivity (usually 0.9) k is the constant 5.67e-8
    For a 1 kW test like Googles, you just need 1 square meter of radiators (assuming two-sided).
    For SpaceX's 175 kW satellites, they will need ~170 square meters of radiators.
    None of this is impossible or even difficult to calculate. That's why I think this is the laziest argument against space data centers. There are so many other more reasonable arguments (like whether they will be economically competitive) but people love to latch on to this one for some reason.
    dekhn · 1790280755 · focus · HN ↗
    To your last paragraph: when opposing something, it makes sense to use the laziest argument first. Only if that doesn't achieve your goals, would you move towards less lazy arguments. I forget what the term for this is, it's generally criticized ("you should just put your strongest arguments first").
    cyberax · 1790282232 · focus · HN ↗
    > For SpaceX's 175 kW satellites
    That's like a quarter of a rack of modern AI hardware. Modern AI datacenters are now in the multi-GW range.
    GMoromisato · 1790282902 · focus · HN ↗
    Okay, so you agree the cooling problem has a solution.
    Now the argument is, what, you can't launch that many satellites?
    echoangle · 1790291740 · focus · HN ↗
    The argument is that launching and operating those satellites is more expensive than just using the same silicon on the ground.
    scottyah · 1790293951 · focus · HN ↗
    I don't think that is the argument, or if so it's an odd one because it's just a fact right now. Instead, people are claiming that it will never be feasible from a purely physics standpoint, which is something debatable.
    echoangle · 1790319721 · focus · HN ↗
    Saying it’s impossible isn’t debatable because it obviously is. What would stop you from a physics standpoint to just launch 100 ISS with Server racks in each? A few kW of installed power in space is already reality, it’s just really expensive.
    GMoromisato · 1790294226 · focus · HN ↗
    How sure are you of that? How much does each satellite cost? How much will silicon on the ground cost in 2030?
    How can anyone be certain of any of those numbers without (a) knowing how the technology will evolve, and (b) doing the math?
    I'm just astounded that people can have such confidence.
    cyberax · 1790294950 · focus · HN ↗
    We know such thing as physics and economics. That dictate the fundamental limits.
    They are showing that space-based DCs only make sense for criminal enterprises. Which probably IS what's going on here.
    NetMageSCW · 1790371981 · focus · HN ↗
    While I really don’t trust Google, I’m not sure I call them a criminal enterprise.
    legohead · 1790282880 · focus · HN ↗
    sure, the physics is solved: we know how radiators work, and we can calculate the area needed, blah blah. but we haven't put sustained-AI-computer systems into space yet - heat is a genuine concern, and personally I'm curious if they are developing something beyond simple radiators. if a ~1m^2 radiator works, great. but if the current test can only run the TPUs for ~15-minute bursts before it has to stop and dump heat, this issue isn't exactly "solved" in my book.
    GMoromisato · 1790283482 · focus · HN ↗
    I don't understand your argument. Sounds like you're saying, "In theory it should work, but what if there are space pixies that keep rebooting the TPUs? What do we do then?"
    The 15-minute figure is what they currently designed for because they have mass constraints. If they wanted a different figure (like forever) they could do it with a larger radiator.
    As you said, the physics is solved! We know exactly how much heat a surface radiates in space--there is literally an equation for it. We know how to cool stuff in space. Will the price be low enough to make a profit? That's the real question. But stop worrying about cooling in space.
    tensor · 1790293024 · focus · HN ↗
    No one thinks we don't know how to cool things in space. Everyone is saying that it's not feasible because you'd have to put too much mass up. Stop derailing the conversation please.
    If you have something useful to contribute regarding how to actually reasonably put up enough mass to make this remotely a feasible idea then please contribute. We have the technology to prevent global warming, and doing that is far easier than this, and yet even that is apparently not feasible for humanity.
    GMoromisato · 1790295038 · focus · HN ↗
    I honestly don't get that. The current SpaceX design is for a ~4 ton satellite with 160 m^2 radiator with peak 250 kW output. You can launch 25 of those satellites on a single Starship launch. 160 Starship launches and you get 4,000 satellites with peak 1 GW compute.
    What's unfeasible about that? SpaceX has already launched 10,000 Starlink satellites. Falcon 9 launches 150 times per year.
    echoangle · 1790291678 · focus · HN ↗
    I think you misunderstand the argument of the people worrying about cooling. I don’t think most people think it’s literally physically impossible, they just think that this will be the thing that makes it economically uncompetitive. It’s a combined argument.
    GMoromisato · 1790294653 · focus · HN ↗
    That may be the argument, but it's a dumb argument. Most of the cost is going to go to chips, solar panels, and launch. Radiators are probably one of the cheapest parts of the satellite: it's a hunk of metal with some pumps for liquid cooling.
    People use that argument because it takes zero thought to make and significant effort to refute.
    echoangle · 1790319790 · focus · HN ↗
    Again you’re missing the combined argument, the cooling will increase launch costs because your hunk of metal will increase the weight of the satellite.
    notahacker · 1790325025 · focus · HN ↗
    People use that argument because one of the chief bullshit arguments put forward by proponents of space based datacentres is "space is cold", when what they actually mean is "space lacks the possibility of using relatively efficient conductive cooling, so we need to expensively launch large amounts of additional mass to use radiative cooling, consider the extent to which we can shield our miles of active cooling loops from impact, and under current plans all of this mass must be replaced on the same cycles as the chips"
    "Space is cold" is classic zero effort to make, significant effort to refute stuff: proponents imply complete nonsense about vacuums being optimal for cooling and anyone who understands why this is a lie to gull retail investors ends up getting bogged down in "sure, but Boltzmann equations proving it's prohibitively expensive doesn't prove it isn't possible"...

Open on Hacker News to reply ↗

Unofficial Hacker News client; not affiliated with Y Combinator.