Well... today I found such a diatribe... sorta of.
In an interview, McCalip says his initial rough calculations a few years ago suggested that data centers in space would cost in the range of 7 to 10 times more, per gigawatt of capacity, than their terrestrial counterparts. "It just wasn't practical," he says. "Not even close." But when Elon Musk began publicly backing the idea, McCalip revisited the numbers using publicly available information about Starlink's and Tesla's technologies and capabilities.So, the Main Smartest Guy in the Room (all the time)© began beating his drum (or other appendage, hehe) for the effort. And, as one might expect, he pushed his LEO Taxi Service in service to in orbit data centers.
What so many ooh and aah over, the reusable lifter, is really Olde Tech. 1 - reusability was demonstrated at least as far back as the SRB for the Space Shuttle, 2 - the landing on a dime from re-entry goes back to at least the Apollo missions landing on the Moon (you can, I expect, find them on youTube; unless, of course, the moon landings were all FAKE). Nothing new here, move on.
So, the author then backtracks bigly when he accounts that these orbiting data centres would be mated to Starlink locations
Bottom line, with some solid but not necessarily heroic engineering, the cost of an orbital data center could be as low as three times that of the comparable terrestrial one. That differential, while still high, at least nudges the concept out of the instantly dismissible category. "I have my particular views, but I want the data to speak for itself," McCalip says.It's not of no significance that The Musk Ox's taxe service is limited to LEO, not geo-whatever very high orbit. What's the diff, you might ask? Well... very high orbit is well above atmospheric friction, so falling out of orbit is a rare occurence. With LEO, on the other hand, falling out of orbit in short order is part of the taxi service bargain.
If you visit the innterTubes you'll find: Starlink sats last 5-7 years and 1 or a couple plummet to Mother Earth daily.
For this illustration, we picked a configuration with an aggregate 1 GW of capacity. The network would consist of some 4,300 satellites, each of which would be outfitted with a 1,024-square-meter solar array that generates 250 kilowatts. The data center on that satellite, powered by the array, might have at least 175 GPUs; McCalip notes that a popular GPU rack, Nvidia's NVL72, has 72 GPUs and requires 120 to 140 kW.Notice, of course, that the author elides the need to replenish the sats. My wild ass guess? Such a host of orbiting data centre component sats could never be fashioned as an on-going concern; The Musk Ox couldn't keep up with the replinishing requirement. And consider two other snags: cooling all those motherfuckers and "wiring" those 4,000 units into a whole data centre. Much easier to do the wiring here on Mother Earth. Ah, but Arizona, data centre ground zero, is on the verge of going waterless. Air cooled computers, mainframe and mini, have been in service for decades, but these AI fashioned ones... nope. Ain't much air up there in space. Cooling is still up in the air. hehe.
Free cooling is perhaps the biggest misconception. Space is cold, but it also has no atmosphere. That means the best heat-removal mechanisms, conduction and convection, are off the table. The only option is radiation. To prevent a chip from overheating in space, a large, costly surface area is required to dissipate the energy and then radiate it.Stupid is as stupid does.
The governing equation for radiative cooling, the only type of cooling available in space, is known as the Stefan-Boltzmann Law. It states that the amount of power you can radiate is proportional to the area of the radiator times its temperature to the fourth power.
No comments:
Post a Comment