Summarizer

It kinda does make sense if you consider that solar panels in space have been used for a very long time (to power satellites). However, getting the electricity they generate down to Earth is very complicated, so you end up having to use it in space, and one of few things that would make sense for that is indeed data centers, because getting the data to Earth is easier (and Elon already handily has a solution for that). However I'm curious how many solar panels you would need to power a typical data center. Are we talking something like a large satellite, or rather a huge satellite with ISS-size solar arrays bolted on? Getting rid of the copious amounts of heat that data centers generate might also be a challenge ( https://en.wikipedia.org/wiki/Spacecraft_thermal_control )...

The plan seems to be for lots and lots of smaller satellites. For inferencing it can work well. One satellite could contain a handful of CPUs and do batch inferencing of even very large models, perhaps in the beginning at low speeds. Currently most AI workloads are interactive but I can't see that staying true for long, as things improve and they can be trusted to work independently for longer it makes more sense to just queue stuff up and not worry about exactly how high your TTFT is. For training I don't see it today. In future maybe. But then, most AI workloads in future should be inferencing not training anyway.

Sending post-compute radio waves to Earth is much safer than sending back TW of power.

Reliably and efficiently transport energy generated in space back to earth, for starters Or let me guess, its going to be profitable to mine crypto in space (thereby solving the problem of transporting the "work" back to earth)

The political issues in space are mostly launch related, right? Once you have the birds up nobody cares about anything except space junk and bandwidth. They're getting experience of solving those with Starlink already. And if you can find a way to put the satellites really far out there's plenty of space - inferencing satellites don't need to be close to Earth, low latency chat stuff can stay on the ground and the flying servers can just do batch. The politics on the ground is much harder. Countries own the land, you need lots of permits, electricity generation is in contest with other uses.

Why would you transfer the energy to earth? The energy powers ai compute = $

Dead on, You can transmit data to and from space and have the compute completed at potentially fractions of the cost.

Not sure why this is downvoted. Much cheaper to transfer data than energy.

Honestly, there's not a lot else I can think of if your goal is find some practical and profitable way to take advantage of relatively cheap access to near-Earth space. Communication is a big one, but Starlink is already doing that. One of the things space has going for it is abundant cheap energy in the form of solar power. What can you do with megawatts of power in space though? What would you do with it? People have thought about beaming it back to Earth, but you'd take a big efficiency hit. AI training needs lots of power, and it's not latency sensitive. That makes it a good candidate for space-based compute. I'm willing to believe it's the best low-hanging fruit at the moment. You don't need any major technological advances to build a proof-of-concept. Whether it's possible for this to work well enough that it's actually cheaper than an equivalent terrestrial datacenter now or in the near future is something I can't answer.

And what happens every time a rack (or node) fails? Does someone go out and try to fix it? Do we just "deorbit" it? How many tons per second of crap would we be burning in the upper atmosphere now? What are the consequences of that? How do the racks (or nodes) talk to eachother? Radios? Lasers? What about the Kessler Syndrome? Not a rocket scientist but 100% agree this sounds like a dead end.

Communication is a well-understood problem, and SpaceX already has Starlink. They might need pretty high bandwidth, but that's not necessarily much of a problem in space. Latency could be a problem, except that AI training isn't the sort of problem where you care about latency. I'd be curious where exactly they plan to put these datacenters... In low Earth orbit they would eventually reenter, which makes them a pollution source and you'd have no solar power half the time. Parking them at the Earth-Sun L1 point would be better for solar power, but it would be more expensive to get stuff there.

> SpaceX already has Starlink. They might need pretty high bandwidth you mean the network that has less capacity than a fibre pair per coverage area?

eh? They're not gonna lay cable in space. The laser links will be retargetable.

How are you doing pci express x16 with lasers without fiber optics? Have you touched data center hardware in your life?

And how much of that power would be spent on high speed communications with Earth that aren't, you know, a megabit or two per second

I grew up on a rural farm in California with a dial-up connection that significantly hampered my ability to participate in the internet as a teenager. I got Starlink installed at my parents' house about five years ago, and it's resulted in me being able to spend considerably more time at home. Even with their cheapest home plan, we're getting like 100 Mbps down and maybe 20 to 50 up. So it's just not true at all that you would have connections that are a megabit or two per second.

That's not what I'm suggesting. The post says "deep space". If you're going to try to harvest even a tiny percentage of the sun's energy, you're not doing that in Earth's orbit. The comparison is a webcam feed from Mars.

That's pretty much a solved problem. We've had geostationary constellations for TV broadcast at hundreds of megabytes for decades now, and lasers for sat-to-sat comms seems to be making decent progress as well.

Those are for video. AI Chat workflows use a fraction of the data.

That's silly on so many levels. 1. the latency is going to be insane. 2. AI video exists. 3. vLLMa exist and take video and images as input. 4. When a new model checkpoint needs to go up, are we supposed to wait months for it to transfer? 5. A one million token context window is ~4MB. That's a few milliseconds terrestrially. Assuming zero packet loss, that's many seconds 6. You're not using TCP for this because the round trip time is so high. So you can't cancel any jobs if a user disconnects. 7. How do you scale this? How many megabits has anyone actually ever successfully sent per second over the distances in question? We literally don't know how to get a data center worth of throughput to something not in our orbit, let alone more than double digit megabits per second.

Grok doesn’t have video as far as I know. I don’t think it’s so absurd. I don’t know how you scale this. But it seems pretty straightforward.

and, of course and inter-satellite comms and earth base station links to get the data up and down. Starlink is one thing at just above LEO a few hundred km and 20km apart, but spreading these around 10s of thousands of km and thosands of km apart is another thing

There's not even a credible way to transfer meaningful amounts of data to and from a deep-space based data center. What good is compute if you can't interface with it? This is where we are now: https://en.wikipedia.org/wiki/Deep_Space_Optical_Communicati... SpaceX may be leading in short-range (few hundred km) space-to-space data transfer but there is a long way to go for terabit/s deep-space links.

I don't know, 10 years seems reasonable for development. There's not that much new technology that needs to be developed. Cooling and communications would just require minor changes to existing designs. Other systems may be able to be lifted wholesale with minimal integration. I think if there were obstacles to building data centers on the ground then we might see them in orbit within the next ten years. I don't see those obstacles appearing though.

> Right now only upsides ... You are missing some pretty important upsides. Lower latency is a major one. And not having to buy land and water to power/cool it. Both are fairly limited as far as resources go, and gets exponentially expensive with competition. The major downside is, of course, cost. In my opinion, this has never really stopped humans from building and scaling up things until the economies of scale work out. > connect to other satellites and earth If only there was a large number of satellites in low earth orbit and a company with expertise building these ;)

I mostly agree with you, but I don't understand the latency argument. Latency to where? These satellites will be in a sun-synchronous orbit, so only close to any given location on Earth for a fraction of the day.