New Mexico bets future on promiscuous supercomputer Can a supercomputer save a state? Former Presidential candidate Bill Richardson sure seems to think so. Richardson has been championing a massive Xeon-based system dubbed Encanto. The supercomputer currently holds the third slot on the Top500 list, which tracks the performance of large computers, and sits at the center of the …
I've been to New Mexico, and it's nice enough if you are into desert (I liked it fine.) But...
a) As was alluded to in the article, self-sufficient in 5 years? I expect it'll be obsolete by then, unless ongoing upgrades are planned. Most machines on the top 500 are off the list within 2 years except a few that were built out and expanded. I simply wonder what demand there is for buying time on a cluster that is 5 years out of date.
b) The bigger issue... This isn't the 1950's. Ya don't walk up to the computer operator and toss them a stack of punch cards or a tape to feed into equipment physically connected with the machine. The two ways I've seen of feeding in info: 1) Internet connection. Sounds ridiculous, but internet2 has 100 gigabits per second on all links and some other research networks have similarly ridiculous speeds. 2) One place I've visited that does astronomical interferometry, they have a crap 56k leased line (making option 1 impossible), so the remote stations send hard disks* via UPS to the central station for processing, then the central station sends them back. As the old saying goes, "Never underestimate the bandwidth of a station wagon full of tapes hurtling down the highway."
*The station I visited actually still uses tapes. They were sent a box of Deskstars, the one guy there pointed out Deskstars are crap, he was told they'd be fine, and on the first run about 75% of them croaked. Apparently the central office never replaced them and just had them use the old reel-to-reel tape drive instead. I can't make something like this up 8-).
"Oh yeah, it needs about $2m a year to power and cool the supercomputer beast a year as well."
Well, shucks, I can tell 'em how to work around that: build the supercomputer in a New Mexico State Park. You see, if you buy a NM Annual Camping Pass (a bargain at $180/yr. for residents), you can stay in any of the state's lovely parks for $4 a night, with - here's the key - unmetered electricity at no extra cost.
OK, it's a maximum of 50 amps per campsite, but that could be worked around with a few extension cords. The big brain's operating costs could be reduced from millions to a paltry $1,640 a year. And whilst crunching numbers, supercomputer staffers would get to enjoy the view of scenic wonders such as Elephant Butte, City of Rocks, Bottomless Lakes... it can't miss!
Andy Baird, writing from Oliver Lee (the guy who shot Frenchy) Memorial State Park, New Mexico
It all sounds much more like New Mexico building a SMARTer Quantum Computer Hub using Encanto Driver Connections, than anything Material to do with the hardware.
It's a Call to programmers to Use the Technology as Software Innovating Social Development and Create Wealth rather than adding more Clutter/Baggage/Overheads. A Virtual Machine which Builds ITself 42 Provide Offspring with Betas Seeding Money Ventures. Future Server Systems with Embedded Future Perfect Present Memory .............. For the Quantum Loop Processing of Enriched Ore Source.
At the atomic/subatomic [astute]/supersubatomic [acute astute/HyperRadioProActive] levels of Quantum State Programming, the QuBits and Bytes in Play are as Real/SurReal as Cosmic Dust in the Sense that with Due Care and Attention to Detail/AAA+ Rated Perfect Diligence, their arrangement/placement Creates a Universal Picture from Artists of the Genre for Universal Artists of the Genre to ReCreate....in their Play Roles.
That would be AIMother of a Machine which the Fed would Migrate to for Inside Feeds. Hell, See the Potential and Invest Appropriately and the Key Access Codes for Remote Proxy Live InterAction are Shared, Guaranteeing Mutually Beneficial Added Value and Greater Worth.
Nice Move, New Mexico ...... One Small Step, AI Quantum Leap Jumper. Virgin Territory too.
The reasoning is simple - supercomputers need a fair bit of tender lovin', which can only be provided by teams of beautiful*, scantily-clad men and women lovingly cleaning fans and replacing failed components.
All of which is quite difficult to do in a spacesuit under zero gravity, not to mention that it's really expensive to send people and material up there.
(*One can dream)
Is actually far from easy. No convection, see; you have to radiate all your heat. Consider things like thermos flasks which use a vacuum layer as part of their insulation. Or double glazing, which stops heat loss through convection as well.
And whilst you get lots of solar flux, potentially all the time if you pick the right orbit, solar cells are still big, delicate and not hugely powerful. And then there are the maintenance callout fees...
It would be more useful to build your datacentre into a hydroelectric dam. Lots of power, and cooling on tap.
"The reasoning is simple - supercomputers need a fair bit of tender lovin', which can only be provided by teams of beautiful*, scantily-clad men and women lovingly cleaning fans and replacing failed components.
All of which is quite difficult to do in a spacesuit under zero gravity, not to mention that it's really expensive to send people and material up there." ......
Surely that is something to be practised in Simulated Space Environment....GroundZero Moon/Mars/Venus. Getting there's the easy bit. Doing something worthwhile though, is QuITe A.N.Other Matter.
But XXXXtraTerrestrial Space Journey Simulation would produce a Source Creative Architecture for Societies which can be Transferred to Earth Systems.
Or in a desert and only use it at night?
If they have a market earmarked there is not much to argue with so long as the contract is signed.
How many New Mexicans will be taken on to maintain it? And why is the running cost going to be 2 million dollars. That seems inordinately high for a computer, or do they include the cost of a solar energy plant?
It actually begins to make sense now. Cheap real estate, free energy in the day and night cooling. Combine that with plenty of available labour.
I imagine a lot of high-tech projects might be attracted to the region if they were offered that background.
Actually getting rid of heat from electronics in space is harder than you think. You have one side of your spacecraft heated by the Sun, and the other side at 3K or some such. And you have a bunch of electronics that gets hot and can't get rid of the heat via convection, only radiation.
It's for this reason that IR telescopes generally carry several years' worth of liquid nitrogen.
Operating costs could be reduced by placing this dinosaur on a cold mountain top! Surrounded by the eternal snows, the cooling costs would be minimal.
As for local jobs creation, I can foresee employment for hundreds of happy New Mexicans either slapping reels of tape onto drives from 1960's James Bond movies, or more modern versions of sneakernet involving USB ports and something similar to the Old Mexican hat dance.
The New Mexican government is to be congratulated on their concern for their constituents.
With projects like seti@home and folding@home now time-tested and demonstrable supercomputing solutions, I would think the days of massive centralised supercomputers are numbered.
Now, the current limitation of distributed computing projects is that they only work if they capture the public imagination, such as the idea of helping to find alien civilisations. Such mundane tasks as city planning and infrastructure logistics aren't likely to induce millions of people to donate their CPU cycles to the cause - boring!
This problem, however, could be alleviated by offering to subsidise part of users' ISP bills in return for running the distributed computing software on their machines. Considering the millions of dollars spent on building and maintaining supercomputers, I'm certain that a cost/benefit analysis would indicate that a subsidy system would be a far cheaper way of effectively using distributed computing in a supercomputing application. If you set it up so that for every X work units processed we'll pay Y dollars toward your next internet bill, you'd find that a lot of people would join in the distributed system who otherwise would have no incentive to do so. Of course, you'd have to impose a limit on the number of clients you allow to join in any given project, based on the computational requirements of the project, the timeframe for completion, and the budget. You'd then have a central website where people could sign up for these subsidised projects on a first-in-first-served basis. This system then means the only cost is the subsidy and central server maintenance, instead of a massive energy-consuming beast of a data centre that costs millions a year to run and maintain.
Good point on the cooling. The bigger point is CPUs, however: space-hardened hardware is, by today's standards, agonizingly slow with IBM's space-certified POWER chips topping out at approx. 200 MHz and Intel, IIRC, currently offering an 80486 for space use. Any current off-the-rack CPUs would be toast within a very short period of time due to all sorts of irradiation that we down here on terra firma are shielded from thanks to our atmosphere plus magnetic field.
So a space supercomputer would have approx. the effectiveness of, say... oh shucks, just go to Walmart and pick up their top-of-the-line office PC. It'll have more oomph and be lots cheaper to operate. Plus, far easier to repair.
The Register 
Biting the hand that feeds IT
Copyright. All rights reserved © 1998–2024
