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

Datacenters in space aren't going to work

taranis.ie

Post reply on HN
Print view View original

211–220 of 447 posts

Re: Datacenters in space aren't going to work

#211

Post by verzali »

GMoromisato wrote: post #185
kaashif wrote: post #165

Earlier quoted context omitted.

> The only one that might apply is cost: if it's cheaper to build data centers on Earth, then space doesn't make sense (and it won't happen). So the only problem left to be solved is that space datacenters would be millions of times more expensive per unit of compute than a ground based datacenter. And cost millions of times more to maintain.

Starlink cost maybe $10 billion. A 100,000 gpu data center costs between $20 and $40 billion to build. Also remember that data centers last for about 5 years; after that the gpus are obsolete. That’s no different than the lifetime of a Starlink satellite.

Starlink solar panels generate at best 200 W/sqm on average. Even with 2.5 million square metres, that is a total of half a gigawatt. And the cost is not to be ignored! Most of the cost of these data centres is in the GPUs themselves, so you need to add that to the cost of building out the constellation. Unless you are arguing that the cost of supporting infrastructure (cooling, power, etc) costs $10bn to support half a gigawatt of GPUs in the typical data centre, then your numbers are simply way off.

Re: Datacenters in space aren't going to work

#212

Post by abalone »

I am skeptical as well BUT on the cooling question, which is one of the main concerns we all seem to have, the article is doing a bit of an apples-to-oranges comparison between the ISS and a cluster of small satellites.

It cites the ISS's centralized 16kW cooling system which is for a big space station that needs to collect and shunt heat over a relatively large area. The Suncatcher prototype is puny in comparison: just 4 TPUs and a total power budget of ballpark 2kW.

Suncatcher imagines a large cluster of small satellites separated by optical links, not little datacenter space stations in the sky. They would not be pulling heat from multiple systems tens of meters away like on the ISS, which bodes well for simpler passive cooling systems. And while the combined panel surface area of a large satellite cluster would be substantial, the footprint of any individual satellite, the more important metric, would likely be reasonable.

Personally I am more concerned with the climate impact of launches and the relatively short envisioned mission life of 5 years. If the whole point is better sustainability, you can't just look at the dollar cost of launches that don't internalize the environmental externalities of stuff like polluting the stratosphere.

Re: Datacenters in space aren't going to work

#213

Post by rgmerk »

cwal37 wrote: post #198
echelon wrote: post #182

Earlier quoted context omitted.

They've always been able to do this. Microsoft was talking about submarine data centers powered by tidal forces in the early 2000s. There have been talks of data centers on Sealand-like nation states. Geothermal ... Exotic data center builds will always be hyped. Always be within the realm of feasibility when cost is no object, but probably outside of practicality or need. Next it'll be fusion-powered data centers.

Commonwealth Fusion Systems called dibs on next last year by saying they’re gonna have a Dominion (Virginia) commercial site up and running in the early 2030s. https://cfs.energy/news-and-media/commonwealth-fusion-system...

Is there a way I can take bets on this not happening? Because I’d sure like to.

Re: Datacenters in space aren't going to work

#214

Post by mangecoeur »

collinmcnulty wrote: post #197

To say the quiet part out loud, I don't think any serious companies have any intention to build a data center in space. There is no benefit in actually trying this. There is however, benefit in saying you'll do it to advance a narrative and distract from the problems terrestrial data centers are facing to an audience that mostly doesn't understand how heat transfer in a vacuum works.

Many of the dumb ideas being hyped in this AI bubble make sense viewed through this lens.

Data centres stirring up opposition? Sell a sci-fi vision that you will move them to Space! And reassure your over-extended investors that the data centre buildout rush you’re committing to isn’t going to get bogged down in protests and lawsuits.

The people hyping this stuff are not stupid, just their real goal (make as much money as possible as quickly as possible) has only a vague relationship to what they claim to be doing.

Re: Datacenters in space aren't going to work

#216

Post by iDon »

Related (posted just 2hours before this article) : https://news.ycombinator.com/item?id=46086833 "Blimps lifting quantum data centers to the stratosphere? (newatlas.com)" "... blimps, to lift quantum computers to the stratosphere. There, at an altitude of about 20 km (12.4 miles), temperatures are in the -50 °C range (about -58 °F) and would be cold enough to allow the qubits to function correctly."

Re: Datacenters in space aren't going to work

#217

Post by ACCount37 »

abalone wrote: post #212

I am skeptical as well BUT on the cooling question, which is one of the main concerns we all seem to have, the article is doing a bit of an apples-to-oranges comparison between the ISS and a cluster of small satellites. It cites the ISS's centralized 16kW cooling system which is for a big space station that needs to collect and shunt heat over a relatively large area. The Suncatcher prototype is puny in comparison: j…

Starlink v2 Mini has about 35 kW of solar power at peak irradiance. 2 kW is quite far from the limit of how much juice we can pack into modern mass produced satellites.

Re: Datacenters in space aren't going to work

#218

Post by ACCount37 »

collinmcnulty wrote: post #196
bouncycastle wrote: post #191

Earlier quoted context omitted.

Recently the USA blew out some some boats in international waters and came back to finish off the survivors, despite thin evidence and no due process, while maintaining that it was legal. If those data centers on ships ever become declared as a 'threat to national security' then they might get the same treatment.

I think GP's point is that an advanced nation-state could just as easily shoot down an orbiting data center as an oceanic data center and that "international space" offers an equally flimsy defense as "international waters" but a much larger price.

Antisatellite weapons are expensive and rare, and also woefully inadequate for dealing with megaconstellations.

If there's one large orbital datacenter, then sure, ASAT is a threat to it. But if it's a dispersed swarm like the Starlink system?

Good luck making a dent in that. You'd run out of ASAT long before Musk runs out of Starlink.

Re: Datacenters in space aren't going to work

#219

Post by ACCount37 »

hedora wrote: post #159
kevdev wrote: post #6

As someone with a similar background to the writer of this post (I did avionics work for NASA before moving into more “traditional” software engineering), this post does a great job at summing up my thoughts on why space-based data centers won’t work. The SEU issues were my first though followed by the thermal concerns, and both are addressed here fantastically. On the SEU issue I’ll add in that even in LEO you can s…

Single event upsets are already commonplace at sea level well below data center scale. The section of the article that talks about them isn’t great. At least for FPGAs, the state of the art is to run 2-3 copies of the logic, and detect output discrepancies before they can create side effects. I guess you could build a GPU that way, but it’d have 1/3 the parallelism as a normal one for the same die size and power budg…

If you're using GPUs, you're running AI workloads. In which case: do you care?

One of the funniest things about modern AI systems is just how many random bitflips they can tank before their performance begins to really suffer.

Re: Datacenters in space aren't going to work

#220

Post by fsh »

alextingle wrote: post #208
fsh wrote: post #204

Earlier quoted context omitted.

A sports car radiator has about that size and dumps 1 MW without boiling the coolant.

A car's "radiator" doesn't actually lose heat by radiation though. It conducts heat to the air rushing through it. That's absolutely nothing like a radiator in a vacuum.

That's the point. Forced air cooling is way more efficient than radiative cooling.
Post reply on HN