Editor’s note: … In this article, we discuss the technical challenges of building an orbital data center constellation: launching all of it, dissipating heat in space, dealing with radiation, and addressing latency issues in orbit. Read part one here.
I find the napkin math interesting, especially putting into light that given expected longevity of such satellites, 5 to 7 years, they will have to do 10 to 42 launches per day. SpaceX will need $1.5 to $10 trillions to make it happen. All of that so the slop machine doesn’t have to run into obstacles like democracy ? So it can destroy communities and the environment freely? What are we doing?
answer: hard.
Space isn’t cold the way cold environments are here on earth. The cold we experience on Earth is because we are surrounded by cold gases. Cold wind and cold air will wick heat away from objects. Because space lacks any atmosphere, all thermal transfer in space is radiative. In comparison to convective thermal transfer via air / atmosphere, radiative thermal transfer is abysmally slow, requiring expansive delicate heat sinks which must be carefully aimed and coated to facilitate thermal egress while impeding solar thermal ingress. If you’ve ever seen a picture of the space station, much of what you may think are solar panels are in fact heat sinks. You can’t just stick a server container in orbit and call the job done.
I see two issues here: get all that stuff into a stable orbit, which would be a way bigger job than building the ISS, which is one of the most expensive projects ever undertaken. Second, it would have to have massive radiators to get rid of the generated heat.
Google and Microsoft already tried underwater data centers to better handle heat dissipation. It wasn’t worth it. Hardware can fail and you need somebody to go in and swap boards and faulty cables. Every complex system has multiple points of failure. The wrong board and the whole container stops working. It was so much pain trying to maintain it under water they all gave up after the Proof of Concept stage.
How are they going to deal with it cost effectively up in orbit? Little nanosat modules? Humanoid robots that barely work today?
Be a lot cheaper (and faster) sending a tech in a little cargo van and a toolbox out to the suburbs of Memphis, Phoenix, or Bakersfield.
So you need SpaceX to increase its launch capacity to somewhere between 20 to 80 times its current rate, completely reimagine the economics of space-based radiators and solar panels, accept that all that shiny new power and cooling CapEx is just as disposable as the GPUs, and you have to adjust the expected use cases and SLAs to allow for nothing being repairable and latency making them unsuitable for many AI tasks.
Yes. “Possible.” I mean, I guess, but we’ve got some powerful Hyperloop energy here. I cannot imagine it ever being more economical than bullying and bribing rural areas with access to a power grid.
And the already predicted heavy damage to the ozone layer from aluminium oxide from re-entering sattelites would be increased a thousand fold so somewhere on that list is learning to live without oxygen or sunlight.
Wouldn’t it still be cheaper to build megastructures on earth in remote and uninhabited places? Like I get you have to build your own power and run your own fiber if you say build a data center the size of Rhode Island in the Yukon or something, but that has to be more cost effective than orbital.
Taking it to orbit creates far more problems than it solves.
I work in this industry and its hard enough to get a GEO/LEO bird in the air let alone a data center…
Not to mention the payload reqs on this. All the radiation shielding and heat management is gonna need a new Sea Dragon-sized rocket.
Great point, also think about the cost in fuel to keep it in LEO orbit. Then what happens when they lose control of it? Space Lab Part 2
Just here to comment that I only pasted the title of the article as the title of the post. I find the content of the article interesting. I’m not actually asking that question, though answer it if you please.
I’d imagine not very hard, but then would it even be worth it? I’m imagining something like in Cyberpunk: an orbital digital prison where one company keeps its enemies’ minds on chips that can be swapped to an entirely different person and overwrite their original personality.
