Hacker News Board Hacker News, browsed like a classic message board
Live data from Hacker News

Data centers in space makes no sense

civai.org

Post reply on HN
Print view View original

911–920 of 1001 posts

Re: Data centers in space makes no sense

#911

Post by fuzzfactor »

indoordin0saur wrote: post #889
zbentley wrote: post #744

Earlier quoted context omitted.

> Data center in space could contribute to the EU's Net-Zero goal by 2050 How unbelievably crass. "Let's build something out of immense quantities of environmentally-destructive-to-extract materials and shoot it into space on top of gargantuan amounts of heat and greenhouse gas emissions; since it won't use much earth-sourced energy once it's up there, that nets out to a win!" Insane.

Blue Origin at least runs its rockets on hydrogen whose exhaust is only water.

Where do they get the hydrogen without putting a load of CO2 into the atmosphere just to manufacture the hydrogen to begin with?

One thing to think about is debt which is not in terms of money.

People are becoming more familiar with "technical debt" since otherwise it comes due by surprise.

With hamsterwheels in space you've got energy debt.

Separate from all other forms of debt that are involved.

Like financial debt, which is only a problem if you can't really afford to do the project so you have to beg, borrow, and/or steal to get it going.

On that point I think I'd be a little skeptical if the richest known person can't actually afford this easily. Especially if he really wants it with all his heart, and has put in any worthwhile effort so far.

Anyway, solar cells are kind of weak when you think about it, they don't produce the high output of a suitable chemical reaction, like the kind that launches the rockets themselves. Which releases so much energy so fast that it's always going to take a serious amount of time for the "little" solar cells to have finally produced an equal amount of energy before a net positive can begin to accrue.

Keeping the assets safely on the home planet simply provides a jump-start that can not be matched.

All other things being unequal or not.

Re: Data centers in space makes no sense

#912

Post by jvanderbot »

rootnod3 wrote: post #798
jvanderbot wrote: post #795

Earlier quoted context omitted.

Unless you want cash-rich AI companies to pay a certain company for weekly / daily lifts during the build-out phase. Eventual crash be damned, make hay while the sun shines

And burn even more energy and fuel than DCs on the ground, great idea :D

Nobody said it was a good idea except his accountant, who doesn't pay the energy bills.

Re: Data centers in space makes no sense

#913

Post by dev_l1x_be »

Dasistmeinname wrote: post #735
dev_l1x_be wrote: post #722

Earlier quoted context omitted.

The Stefan-Boltzmann Law tells us that radiative power scales to the fourth power of temperature (T^4). While terrestrial cooling is largely linear and dependent on ambient air/water temperature (the "wet-bulb" limit), a radiator in space is dumping heat into a 3-Kelvin sink. That thermal gradient is massive.

This is misleading: - A radiator only “sees” 3 K if it’s perfectly shielded from the Sun, Earth albedo, and Earth IR. In Earth orbit you can easily get hundreds of W/m^2 incident; without sunshields the net rejectable heat is greatly reduced. - You have a "massive" advantage only if the radiator is allowed to run very hot: At 300–310 K with \epsilon \approx 0.9: about 400–500 W/m^2. Effective "radiative heat transfer…

Yeah I took the best case scenario in space, I did not account for anything else. I imagine the space DC be like two sides, one is pointing towards the sun and being a solar panel the other is a cold radiator radiating heat into the void. I am not sure how feasible this is.

Re: Data centers in space makes no sense

#914

Post by tpurves »

Closest thing to a usecase I can think of are all military in nature. There's an argument that AI in space only begins to make any sense if the data source you are processing is also originating in space, and you need (for reasons) to run models on it in space before downlinking the results.

Maybe there's a very latency sensitive need to send realtime targeting information to a tomahawk missile in flight? But it's also too bandwidth, compute or cost intensive to send a firehose of raw spy-satellite data to a disposable one-way attack munition?

The data centers in space are actually for the spy satellites to use. That's all I got for practical applications.

Re: Data centers in space makes no sense

#915

Post by tim333 »

beloch wrote: post #115

I would not assume cooling has been worked out. Space is a vacuum. i.e. The lack-of-a-thing that makes a thermos great at keeping your drink hot. A satellite is, if nothing else, a fantastic thermos. A data center in space would necessarily rely completely on cooling by radiation, unlike a terrestrial data center that can make use of convection and conduction. You can't just pipe heat out into the atmosphere or build…

>I would not assume cooling has been worked out

On the other hand Starlink has several thousand satellites up there using solar power to run processors and cooling them with radiators so it's not totally new technology.

Here's a Musk tweet linking some analysis https://x.com/elonmusk/status/2013676764099199156

Re: Data centers in space makes no sense

#916

Post by dev_l1x_be »

papercrane wrote: post #729
dev_l1x_be wrote: post #722

Earlier quoted context omitted.

The Stefan-Boltzmann Law tells us that radiative power scales to the fourth power of temperature (T^4). While terrestrial cooling is largely linear and dependent on ambient air/water temperature (the "wet-bulb" limit), a radiator in space is dumping heat into a 3-Kelvin sink. That thermal gradient is massive.

The thermal gradient in space is meaningless because there is hardly any matter to dump the energy into. This means you are entirely reliant on thermal radiation. If you look at the numbers given by Stefan-Boltzmann law you'd see that means to radiate a significant amount of energy you need a combination of a lot of surface area and high temperatures. This means you need some sort of heat pump. For a practical exampl…

The ISS comparison is a bit of a category mismatch. The EATCS is complex because it’s a life-support system that must keep humans at exactly 22C (295K) while managing ammonia loops in a manned environment.

Computers aren't humans. High-performance silicon can comfortably operate at a junction temperature of 80C to 90C (approx. 360K). Because of that T^4 relationship, a radiator at 85C rejects nearly double the heat per square meter than a radiator at 20C, unless I miss something.

So this makes it a bit more nuanced.

Re: Data centers in space makes no sense

#917

Post by aftbit »

Terr_ wrote: post #31

I'm convinced that >30% of this comes from ideas leaking out of fiction such as like Neuromancer , and percolating through the minds of wealthy people attracted to some of the concepts. Namely, the dream of being a hyper-wealthy dynasty, above any earthly government, controlling an extraterritorial Las Vegas Fiefdom In Space. (Which in the book, also hosted a powerful AI.) Then they work backwards, trying to figure o…

Some of it might come from True Names by Vernor Vinge too, where the cheapest place to buy bulk compute is giant satellites orbiting on geosync orbit. This became possible because they were developed early on for communications, then obsoleted by Starlink-like LEO constellations, so they have a ton of spare undesirable high-latency capacity to sell. In this fictional world, ground-based fiber optic networks never happened, at least not at the kind of scale they did in the real world.

But because that's fiction, Vinge can just handwave away all the hard engineering problems for sci-fi flavor.

Re: Data centers in space makes no sense

#918

Post by zippyman55 »

I loved when we would have an HPC server crash and we would analyze the issue and the vendor would respond, "probably a neutrino strike". I can't imagine this being more complex, given all the constraints of being in an orbit. The hardware depreciation, and the need to eliminate charge buildup and excess heat. Placing servers in the Arctic seems a better solution, at least the cooling issue is resolved.

Re: Data centers in space makes no sense

#919

Post by dev_l1x_be »

shadowgovt wrote: post #728
dev_l1x_be wrote: post #722

Earlier quoted context omitted.

The Stefan-Boltzmann Law tells us that radiative power scales to the fourth power of temperature (T^4). While terrestrial cooling is largely linear and dependent on ambient air/water temperature (the "wet-bulb" limit), a radiator in space is dumping heat into a 3-Kelvin sink. That thermal gradient is massive.

Stefan-Boltzmann is about absolute, not relative temperature. When one does the math on the operating temperatures of regular computing equipment that we use on Earth, how much heat it generates per watt, and how fast it would need to sink that heat to allow for continuous operation, one gets surface areas that are not impossible, but are pretty on the high end of anything we've ever built in space. And then you have…

You’re correct that Stefan-Boltzmann uses absolute temperature (K), but that only reinforces the advantage of moving the "hot side" of the gradient up. If you increase your radiator temp from 300K (standard Earth ambient) to 360K (hot silicon), your radiative efficiency doesn't just go up by 20%—it nearly doubles.

The Solar Load is Directional: Unlike a terrestrial atmosphere where heat is omnidirectional, space allows for "shadow engineering." A simple multi-layer insulation (MLI) sunshield can reduce solar flux by orders of magnitude. We do this for the James Webb Space Telescope to keep instruments at 7K while the sun-facing side is at 380K. For a data center, you don't need 7K; you just need to keep the "dark side" radiators in the shade.

Re: Data centers in space makes no sense

#920

Post by paulryanrogers »

therein wrote: post #904
slashdev wrote: post #894

Earlier quoted context omitted.

As an engineer of the software variety, logically heat dissipation would seem like a difficult problem. But SpaceX has lots of real engineers who are very smart. I’m certain they ran the math on it. Which is more than you or I have done. If they say it can be done, I’m inclined to believe them.

[flagged]

I think it's more an appeal to (expert) authority
Post reply on HN