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

Orbital Data Centers: Why the Hype Outpaces Reality

spectrum.ieee.org

Post reply on HN
Print view View original

21–30 of 67 posts

Re: Orbital Data Centers: Why the Hype Outpaces Reality

#21

Post by chuckadams »

rbanffy wrote: post #17
chuckadams wrote: post #13

Earlier quoted context omitted.

Train-in-a-vacuum-tube isn't even restricted by the laws of physics, and it doesn't have to be a perfect vacuum anyway, just low pressure. Data Centers IN SPAAAAAACE have the teensy little problem of cooling that you can't just brute-force a solution to. Those giant things dangling off the ISS that make up most of its footprint aren't solar panels, they're heatsinks, and they still have issues managing heat.

Well... If the launch cost is low enough you can just pack a radiator large enough. Good thing this might drive the development of a wider Starship and Ultraheavy booster (or a Superheavy-Heavy where three Superheavy rockets boost a bigger, heavier Starship). Eventually we can get to the Comicallyheavy booster.

Perhaps, but I don't see the costs getting anywhere near being worth it. You'd need a space elevator or to manufacture the sinks IN SPAA— er, in space. The latter is theoretically possible but still extremely remote, the former still requires unobtanium.

Re: Orbital Data Centers: Why the Hype Outpaces Reality

#22

Post by joloooo »

NishanStepak wrote: post #2

The advantage is solar power in space. They will never stop getting sunlight.

There's a very narrow band of sun-synchronous orbit with consistent enough sunlight to make orbital data centers viable. Even with a tight constellation of 500,000 to 1,000,000 satellites packed into that band, it still doesn't make economic sense compared to what terrestrial data centers can do. At 500,000 satellites you're looking at maybe 50 GW of solar power. Then factor in the lifespan of LEO at ~5 years you're looking at hundreds of satellite launches daily to maintain that infrastructure.

Now work is being done for increasing solar access (Starcatcher) and others are working on improving LEO refueling and repair capabilities, but I would say we're decade+ away from establishing any true compute infrastructure in orbit.

Re: Orbital Data Centers: Why the Hype Outpaces Reality

#23

Post by aeternum »

CuriouslyC wrote: post #8
aeternum wrote: post #6

Earlier quoted context omitted.

Good thing Elon's companies have a history of moving engineering impossibilities from impossible to slightly late. Remember when globally competitive electric cars, re-usable boosters, catching a rocket with chopsticks, playing a fps game via a brain implant, and maintaining a satellite constellation at 480km LEO were also impossible?

There's a difference between "infeasible given materials and manufacturing capabilities of the time" and "infeasible as a scalable solution due to the laws of physics"

What law of physics do orbital data centers violate? You must have learned some strange physics.

Re: Orbital Data Centers: Why the Hype Outpaces Reality

#24

Post by CuriouslyC »

aeternum wrote: post #23
CuriouslyC wrote: post #8

Earlier quoted context omitted.

There's a difference between "infeasible given materials and manufacturing capabilities of the time" and "infeasible as a scalable solution due to the laws of physics"

What law of physics do orbital data centers violate? You must have learned some strange physics.

note the scalable in there. Even under best case assumptions, it's only a good fit for edge coprocessing where latency is super important. You're not going to power the world's compute with a self healing Dyson sphere (but if you think that, I've got something to sell you).

Re: Orbital Data Centers: Why the Hype Outpaces Reality

#25

Post by grim_io »

aeternum wrote: post #23
CuriouslyC wrote: post #8

Earlier quoted context omitted.

There's a difference between "infeasible given materials and manufacturing capabilities of the time" and "infeasible as a scalable solution due to the laws of physics"

What law of physics do orbital data centers violate? You must have learned some strange physics.

How about the physics of space radiation and radiative cooling.

The guy has to rent out his terrestrial data centers because he can't make use of them.

We probably don't need to pollute space with dead GPU's.

Re: Orbital Data Centers: Why the Hype Outpaces Reality

#26

Post by aeternum »

chuckadams wrote: post #13
automatic6131 wrote: post #9

Earlier quoted context omitted.

Almost no one said those were impossible, just hard. This is completely different. Rather like... a train in a vacuum tube hard. Definitely harder than making a subway with autonomous trains under a modern city. And much harder than being a third rate AI lab, though Elon did hit that target perfectly.

Train-in-a-vacuum-tube isn't even restricted by the laws of physics, and it doesn't have to be a perfect vacuum anyway, just low pressure. Data Centers IN SPAAAAAACE have the teensy little problem of cooling that you can't just brute-force a solution to. Those giant things dangling off the ISS that make up most of its footprint aren't solar panels, they're heatsinks, and they still have issues managing heat.

Energy radiation scales T^4 so physics is really on your side here. If you can engineer GPUs to run a little hotter you get significant decreases in radiator size required.

Re: Orbital Data Centers: Why the Hype Outpaces Reality

#27

Post by rbanffy »

chuckadams wrote: post #21
rbanffy wrote: post #17

Earlier quoted context omitted.

Well... If the launch cost is low enough you can just pack a radiator large enough. Good thing this might drive the development of a wider Starship and Ultraheavy booster (or a Superheavy-Heavy where three Superheavy rockets boost a bigger, heavier Starship). Eventually we can get to the Comicallyheavy booster.

Perhaps, but I don't see the costs getting anywhere near being worth it. You'd need a space elevator or to manufacture the sinks IN SPAA— er, in space. The latter is theoretically possible but still extremely remote, the former still requires unobtanium.

It all depends on how far you are willing to go to make your server invulnerable to a police raid.

Re: Orbital Data Centers: Why the Hype Outpaces Reality

#29

Post by ryandvm »

root-parent wrote: post #3

Data Centers in Space are a practical engineering impossibility, as well as making no economic sense. Engineering and the laws of physics get on the way. Just because Scott Manley refuses to call that out, so he can do another eight videos about it, don´t stop listening to somebody with the feet on the ground: "Orbital Data Centers: Spacecraft Constraints and Economic Viability" - https://arxiv.org/abs/2604.27197 "Ho…

Nobody in this industry with a modicum of applicable expertise believes that orbital data centers make sense financially. Like colonizing Mars and most of Elon's pipe dreams, it's not the stated goal they believe in, it's getting fabulously wealthy from fat government contracts along the way.

Once you understand this about Musk, you realize that everything he is involved with works that way.

Re: Orbital Data Centers: Why the Hype Outpaces Reality

#30

Post by dreamcompiler »

aeternum wrote: post #23
CuriouslyC wrote: post #8

Earlier quoted context omitted.

There's a difference between "infeasible given materials and manufacturing capabilities of the time" and "infeasible as a scalable solution due to the laws of physics"

What law of physics do orbital data centers violate? You must have learned some strange physics.

Thermodynamics.
Post reply on HN