cross-posted from: https://sh.itjust.works/post/24946971
TL;DW:
Does It Make Sense To Put Data Centers In Space?
At some point in the future, yes.
Can They Really Cost Less To Operate?
In theory, yes.
Scott expresses concerns that current startups have not adequately addressed some of the practical challenges, such as cooling.
Up there the sun always shines (assuming a convenient sun synchronous orbit), so you have access to uninterrupted solar power. That’s the only advantage I can think of. You’re going to need a lot of solar panels though, and even more radiators to radiate all the heat away. And a number of other disadvantages.
Considering some data centers basically need a power plant to run, the amount of solar panels needed would be insane. Plus, unless you keep the data center between the planet and the sun, you still have shade.
There is such a thing as Sun-synchronous polar orbits (an example).
Neat
The way I understand is that they would use a sun-synchronous orbit, more specifically a dawn/dusk orbit, which places the satellite over the terminator between day and night meaning it always directly sees the sun. But yes, it would need an insane number of solar panels. What’s more, data centers don’t just need power, they also need cooling. So there would also need to be an insane amount of radiators (in space it’s very hard to get rid of heat).
All in all the only advantage I can see is not much of an advantage, if at all, especially compared to all the drawbacks.
I’m not sure how that’s even an advantage, unless there is some scenario where you want the data center to continue after the end of civilization. Providing reliable power on Earth already has known solutions and much much cheaper
I suppose you want some sort of caching proxies to make the internet somewhat useable from any bases farther out than low orbit but even then it would be much cheaper and easier to host them at those bases.
So we need to reboot a server / swap a drive. Who’s on call?