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

241–250 of 447 posts

Re: Datacenters in space aren't going to work

#241

Post by rsynnott »

mangecoeur wrote: post #214
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 mon…

At the point, it's beginning to feel a bit like the 419 scam (where you make the details deliberately absurd so as to ward off people inclined to be sceptical early, leaving you with only the easiest marks.) SMRs! Data centres in space! "phD level AIs".

Re: Datacenters in space aren't going to work

#242

Post by adastra22 »

ReptileMan wrote: post #5
yardie wrote: post #4

I asked Google for more information about AI datacenter in space. This was the first sentence, 'AI data centers are being developed in space to handle the massive energy demands of AI, using solar power and the vacuum of space for cooling.' > After laughing at "the vacuum of space for cooling" I closed the page because there was nothing serious there. Basic high school physics student would be laughing at that senten…

You can radiate the excess energy away on the non-sun facing part. In theory.

No, you can't. You need to radiate away all the heat being received from the sun facing half, AND excess heat from the compute. Even in theory, the non-sun-facing part doesn't give you any benefit. It's already part of the system that accounted for the temperature of the sun-facing side.

Re: Datacenters in space aren't going to work

#243

Post by adastra22 »

adamwong246 wrote: post #30
morcus wrote: post #18

Earlier quoted context omitted.

The Moon doesn't have a magnetic field, though, so the second half of the article discussing difficulties due to radiation would still apply, right?

We will need to develop very robust, space-worthy electronics eventually. We can't rely on natural magnetic fields forever.

We have robust, space-worthy electronics. They're discussed in the article. You just can't get SOTA performance from them, because of fundamental physics-driven compromises.

Re: Datacenters in space aren't going to work

#244

Post by StephenHerlihyy »

It’s not about putting data centers into orbit. It’s about the cost-yield inversion to data centers cooling infrastructure that happens at terawatt scale. All things being equal - a chilled circuit performs better and produces less heat than a hot one. There is a high up front cost to pre-cooling but if you can get in the -60C range, and stay there, you can increase performance and cut energy costs.

When they say data centers in space - they mean data centers you can’t get to because they are flooded with ultra cold dielectric fluid and it costs tens of millions of dollars to bring them back up to human temperatures.

Right now it’s not worth the hassle. At terawatt scale it’s almost mandatory.

When you walk down that line it’s pretty close to putting them in space. No access. Super cold. No air. Tiny, insulated capsule. Thermal management hell. They’ll be buried in mines though, not launched into orbit.

It’s just corporate propaganda to simplify an otherwise insane situation.

Re: Datacenters in space aren't going to work

#245

Post by KaiserPro »

jillesvangurp wrote: post #227

There are lots of reasons why keeping data centers on the ground might be cheaper but the article seems to be skipping over a few things. 1) ISS is about 30 years old. It's hardly the state of the art in solar technology. Also, it's much easier to get light to solar panels far a larger part of the time. Permanently in some orbits. And of course there is 0% chance of clouds or other obstructions. 2) We'll have Starshi…

1) ISS is about 30 years old. It's hardly the state of the art in solar technology.

Domestic solar panels are heavy, and dont need to deal with hypersonic sand blasting. even at that height, you are in shadow every 90 minutes.

> 3) The article complains about lack of bandwidth. Star Link serves millions of customers with high speed, low latency internet via thousands of satellites.

Right. First power and heat are a massive pain to deal with. You need megawatts to run a datacentre. A full rack of GPUs (48u, 96 GPUs) is around 40-70kw. It also weighs a literal ton.

You also need to be able to power that in the time when you are in darkness. BUT! when you are zooming around the earth every 90 minutes, you can't maintain a low latency connection, because the distance between you and the datacentre.

That means geostationary, as that solves most of your power issues, but now you have latency and bandwidth issues. (oh and power, inverse square law and bandwidth are related)

> 5) Learning effects

Great, but it gets us nothing.

Re: Datacenters in space aren't going to work

#246

Post by craighay1 »

Really the only potential upside to this, and it's a niche one, is for time or security sensitive compute tasks where the raw data originates in orbit. Every happens over inter satellite link and there's no downlink to Earth until the end of the process (downlink is still a problem as Earth is mostly ocean and wilderness)

E.g. one satellite's wide area sensor payload is processed and "potential wildfire detected". The result is passed to another satellite with finer grained sensing capabilities which is due to pass over in the next X minutes which then tees up a capture.

Re: Datacenters in space aren't going to work

#247

Post by khalic »

jillesvangurp wrote: post #227

There are lots of reasons why keeping data centers on the ground might be cheaper but the article seems to be skipping over a few things. 1) ISS is about 30 years old. It's hardly the state of the art in solar technology. Also, it's much easier to get light to solar panels far a larger part of the time. Permanently in some orbits. And of course there is 0% chance of clouds or other obstructions. 2) We'll have Starshi…

Can you please tell me your credentials compared to someone who actually built material that went into space? Like the author of the article

Re: Datacenters in space aren't going to work

#248

Post by eru »

rsynnott wrote: post #241
mangecoeur wrote: post #214

Earlier quoted context omitted.

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 mon…

At the point, it's beginning to feel a bit like the 419 scam (where you make the details deliberately absurd so as to ward off people inclined to be sceptical early, leaving you with only the easiest marks.) SMRs! Data centres in space! "phD level AIs".

You can short the publicly traded companies that do this.

Re: Datacenters in space aren't going to work

#250

Post by qayxc »

GMoromisato wrote: post #189
phsau wrote: post #166

Earlier quoted context omitted.

>Almost every single objection to data centers in space has already been overcome at a smaller scale with Starlink Did you not read the article? It had many objections that make it clear datacenters in space are unworkable...

Starlink is already a small data center! It has power, radiators, and compute! It needs to be scaled up, but there is no obstacle to that (at least none that the article mentions). The only valid objection is cost, but space prices keep dropping and earth prices keep rising.

> Starlink is already a small data center! It has power, radiators, and compute!

It is not. This is like saying your phone is already a small data centre. While technically true, we're not talking about the same scale here. StarLink's compute power is a tiny fraction of a modern data centre GPU/TPU. Most of the power budget goes into communication (i.e. its purpose!).

Post reply on HN