A datacenter in space might be the silliest idea ever proposed.

I like "litho braking event". 😂 Yeah, having a few hundred pounds of plutonium, dispersed across say 500 square miles might not be an ideal end to the project. Hmm... I wonder what happened at Tunguska...? I don't think it's ever been completely solved...
 
blackbird9 said:
Self-driving is a very tough nut to crack. Yes, they've all failed miserably so far, thankfully. I don't want a car without a steering wheel, thank you.
Click to expand...
I don't want a car that can be turned off remotely if I decide not to pay my "use tax" or one where they can upload my driving performance data without my permission. That's one reason the last truck I bought was the last model year without a cellular transponder in it.
 
rrrrrrr
kent_dorfman766 said:
I don't want a car that can be turned off remotely if I decide not to pay my "use tax" or one where they can upload my driving performance data without my permission. That's one reason the last truck I bought was the last model year without a cellular transponder in it.
Click to expand...
Neither do I, but it is increasingly hard to get such cars which are still reliable and road worthy.
 
Phishfry said:
Which seems logical. Some space telescopes probably require huge bandwidth. Get too many and overwhelm systems.
Solution. Process stream in space with AI Datacenter and only relay to earth what is found. Pre-processing streams.
Click to expand...
This is one of the old arguments for fractionated satellites. Except then it wasn't called "AI datacenter", it was called "image processing".
 
until they solve the cooling problem, there isn't going to be any datacenters in space. No convection in vacuum, and no gravity means that cooling will be more costly than on earth (with current cooling solutions at least).
 
I'm hanging on to my 12-year old diesel as long as I can keep it running. Hoping for another 10 years or so. Mostly depends on spare parts availability.
 
If we ever get real superconducting computing, that may be something that would work in space, provided you can keep it cool. It's much more power efficient than standard chips, this page talks about power savings of the order of 500 times. It looks like it's all still in primary research right now. My guess is, Elon or whoever else has been proposing this, is thinking of a 20 to 50-year timeline, and what they are talking about doing right now is an early pilot study, using current technology. You can still do a lot of useful work with the primitive 'model-T' level technology we have today, and scale it as the compute tech evolves over time.

Superconducting computing - Wikipedia

en.wikipedia.org en.wikipedia.org
Well, who knows what they've got cooking away in the labs at LLNL and Los Alamos (and Area 51...) they're hardly likely to tell us if they have something like high-temp superconducting microcircuits working at this stage. Imagine a simple cpu like an ARM, clocked at 700 GHz and running on 1/500th of the power of a current generation chip. I guess having an in-orbit 'data center' would give you a test bed to test something like that. Or you could site a thing like that at the bottom of a dark crater on the moon, which never gets the sunlight and is permanently cold, and have some variant of the recently demonstrated IR laser datalink from it back to earth, via relay satellites.

There is also the thought that if the west doesn't do it, someone else will.
 
ralphbsz said:
Some of the large facilities have spectacularly exploded, which makes for very beautiful fireworks
Click to expand...
That's why I like the idea of Flow batteries for grid storage.

Flow battery - Wikipedia

en.wikipedia.org en.wikipedia.org
They can use Anolytes and Catholytes as benign as Iron salts in aqueous solutions, and their capacity can be scaled by adding larger tanks of those liquids. The Iron redox types are quite mature now and some are deployed in grids in California, UK and etc.
 
You must log in or register to reply here.
Back
Top