Cray T94 Supercomputer on Ebay | Hacker News Reader

Cray T94 Supercomputer on Ebay

ebay.com

104 pointsfrozenport56 comments

"I am liquidating some of my personal collection of unique stuff and my Cray T94 Super Computer must go. What you see in the photos is what I have and what you get. I know that you need more than what I have to make it work. You should consider inspection the item to make sure it is what you want as it is sold as is with no returns."

Open articleView on HN

Apparently, the Cray T90 in the 32-processor configuration had 360GBps of shared memory bandwidth - this is still many times above a shared memory configuration that you can get on the desktop today. Of course, supercomputers have largely moved on from shared memory systems to clusters, and have larger aggregate bandwidth - but shared memory still has its uses. Just a point for all the comparisons to smartphones and what not.

EDIT: just to expand on this a bit, this means that there are workloads that these old-school supercomputers will run much faster than a modern high-end desktop. This particularly applies to workloads which have a lot of shared memory access at random locations - a very difficult case for the modern systems, which depend on high cache hit rates. Also, the GFLOPS ratings of these supercomputer processors are in many ways more real than the ratings of commodity processors, which depend on the pipelines being filled in very specific ways. So no, you can't replace this system (at least the 32-processor version) with a smartphone or even a desktop. Which is not to say it would be cost-effective at 39 million of 1996 dollars.

Could you expand upon the types of applications that benefit from what you describe? I'm guessing simulations, but would love to hear more details.

Most simulations can actually be parallelized in such a way that most memory accesses are local - you can divide the object or system in such a way that various parts don't communicate with each other too much, and map that to a cluster computer. But there are some exceptions in which parts of a simulated system can affect each other remotely. For example, in a lot of nuclear simulations, particles produced in one part of the system can very quickly travel to other parts; you have a problem when you have particles in your system which operate on different time scales, e.g. neutrons, heavy nuclei, and photons. This is a big reason why the DOE and nuclear laboratories liked these supercomputer systems.

Another case which I saw personally was a simulation of a part of the visual cortex of the brain; you had neurons which were connected to their neighbors, but you also had a bunch of connections to far-away neurons, and the bandwidth between processors which simulated different parts of the cortex became a limitation (and the huge supercomputers which had the bandwidth were (a) expensive, and (b) had relatively slow processors for the number crunching in each region).

Except in this case, I found that the physical delay which existed on a long connection between neurons allowed us to buffer the messages and send a notification about the whole train of impulses, effectively compressing the data. Together with some other simple changes, the simulation ran 10 to 100 times faster, and could use clusters instead of supercomputers.

In general, there are not that many cases in which you really can't get rid of the requirement of fast non-local memory access; if there were, these supercomputers wouldn't have died out. But they were useful in some cases, and were also good for freeing people from thinking about how to localize their memory accesses - this speeded up development.

I helped sell a Cray on eBay once, in 2000. I remember offering to let the Pittsburgh Supercomputing Center use my eBay account, because I had hundreds of karma from selling Magic cards.

The coolest part was a decade later, I ran into the guy that bought it... on Hacker News. And that guy was Steve Blank, who turned out to be a hero of mine.

http://steveblank.com/2009/11/19/closure/

>"I'm ecstatic," Blank said. "In 50 years, people are just going to go, 'That was the pinnacle of military computing. These machines are going to be as important as the first PC or the first minicomputer.'" ... Regardless of where the machine ends up, Blank said he plans to keep it in working condition. He said he was arranging with Cray and the supercomputing center to ship the machine in a climate-controlled environment.

And then...

>It sat in my barn next to the tractors and manure for five years. I had the only farm capable of nuclear weapons design. Cray called two years ago and bought it back for parts for an unnamed customer still running one.

That's kind of sad. But he did help ensure another Cray remained in working condition, so I guess that's a positive aspect of the story!

For reference, most smartphones made within the last year or so have superior specs than the Cray T94.

The T94 has a peak theoretical performance of about 8 GFLOPS, 1 gig of ram, half a meg of L2 cache and an SSD up to 4 gigs in size.

http://www.craysupercomputers.com/downloads/CrayT94/CrayT94_...

Compare this to the Galaxy S III, which has at least 1 GB of ram, 64 GB of flash storage, up to 8 MB of L2 cache per core (4x), and nearly 20 GFLOPS of real world performance (on the GPU).

In 15 years time, someone will sell the Galaxy S III on ebay, and people will compare it to the mini processor found in their coffee mug.

Vector supers were bandwidth machines though. The memory is SRAM and a 8 GFLOPS T94 probably() has about 100 gigabytes/sec of theoretical memory bandwidth. Compare that to 6 GB/sec theoretical bandwidth of LPDDR3.

) Cray lists the fully loaded 32-cpu T90 has doing 800 GB/sec, a 8 GFLOPS T94 has 4 CPUs

From Wikipedia:

The system chassis weighs ten tons, contains four tons of fluorinert coolant, and is approximately the shape and size of a very large chest freezer

And yet people complain about their mobile phone not being thin enough. It's amazing how far we've come in only 15 years.

I can't help my cynicism, at least Crays were useful.

Mobile phones are pretty useful too.

It's got me wondering what the market value of fluorinert is.

I quote it: "Note to Laser Technicians : Don't let this fluid drip. It costs like Gold!".

http://www.parallax-tech.com/fluorine.htm#price

Does this unit come with the 4 tons of coolant? According to your link, the fluorinert itself is worth over a million dollars already.

From Cray:

The CRAY T94 mainframe is 5 ft wide, 3 ft deep, and 4 ft high (1.5 m x 1 m x 1.5 m). It has a dry weight of 2000 lbs (909 kg).[1]

So only 120 times thicker than an iPhone 5, but 8,000 times more massive.

[1] https://cug.org/5-publications/proceedings_attendee_lists/19...

I would love to know who is selling this. There are lots of old-looking racing cars in the background of the photo, suggesting someone with a serious hoarding instinct and the bankroll to support it!

Iron Man (was my first thought ;)

Bruce Linsmeyer:

http://www.avonaero.com/

http://www.vdsracingengines.com/aboutus.asp

Nice Lotus 56 in the background. This is a turbine-powered vehicle. Interestingly enough, the seller has some jet engines/turbines for sale as well.

http://www.avonaero.com/STP%20CAR.htm

Thank you for sharing. Apparently the white car(Howmet) is quite a gem and also turbine-powered. For those interested in learning more: https://en.wikipedia.org/wiki/Howmet_TX

You can build your own Cray (of a different model) using FPGAs.

(http://www.chrisfenton.com/homebrew-cray-1a/)

I think he'd be interested in talking to anyone with Cray documentation or software.

We have come a long way in a short time. But I do miss the outlandish design of the Crays, and the blinking lights of Thinking machines. A bunch of racks just isn't the same.

340 volts at 100amps.

Wow, that's...uhh...impressive. I assume most people will just buy it to put in their lobby as an expensive couch.

edit

I was mistaken. The T94 doesn't have the same floor plan as the XMP.

Perhaps it'll make a good beer dispenser then.

Really gives you some context of how far we've come in so little time, eh?

I sometimes find it sad that pieces like this T94 wind up as museum pieces or conversation pieces, but it's just one of the casualties of our rapid rate of progression. I mean, the Qualcomm Snapdragons in cell phones are probably more powerful, and they run off a battery worth a couple thousand mAh.

Buy and hold; Cray just might buy it back from you in a couple years - http://steveblank.com/2009/11/19/closure/

Unless of course this is the machine Cray was buying parts for, and now there's no one left that uses them.

That is a bit of history, not as attractive as the X or Y-MP (to my taste) but still a neat machine. Folks have been comparing it to a smartphone today, which is inaccurate, it would be better to compare it to the GPU in a smart phone where the GPU shader engines don't have nearly as rich an instruction set as the CLUs of the T94. (and the T94 only had 4 CPUs rather than 16 or 32) But a amazing I/O bandwidth of 8 GBytes second was pretty cool. This thing would have been capable of an awesome render of most 3D scenes.

I like how they posted a photo of the power requirements, just so that there be no misunderstandings on whether one can plug it in and fire it up.

You better be able to handle the heat dissipation, too :-)

Look at the design on that thing. It just makes me want to weep when I see a rack full of "SuperMicro's".

And how well-built it is! Look at the construction: all-machined bolts and panels and stuff. Looks like something you'd find in the avionics bay of a spaceship. I love how the panels have milled labels, too: "Common Memory", etc. Looks like it'd sing "Daisy, Daisy" as you opened it up. If I had the means I'd pick this thing up and put it in my living room in a heartbeat. Alas.

Any idea how this would compare in power and capability to today's modern desktop machines?

It compares to smartphones, a modern desktop machine (say, a Core i7 w/ 8 GB of RAM and a 512 GB SSD) would leave this thing in the dust.

Cool. Here's another rarity. A Ma 128 prototype [1] going for almost US 100k.

[1] http://www.ebay.com/itm/RARE-PROTOTYPE-MACINTOSH-128k-COMPUT...

Asking price is different from market price.

Is it? How many of these are currently on the market?

Asking price is what you think someone will pay for it.

Market price is what someone will actually pay for it.

Interesting to see the SAMSUNG sticker on the CRT even back then.

Add to this the Sony floppy "innovation" "magic", and it turns out, that AAPL is just a branding outfit, as The Register never misses to point out.

Lots of folks here rightly bring up how this compares to a smartphone. I think this shows how brutal being in the supercomputer biz is.

Someday I want to see the Hacker Dojo buy an old Cray or IBM mainframe and set it up for people to run random jobs on. Not for any practical purpose, just so people can get a feel for what the old systems were like.

The MoMA should buy it

I see from Wikipedia that the T94 is the 1-4 processor version. How does this compare to a current i5/i7 processor?

The 32 processor version cost $39 million.

Not quite horrifically, but not well. Wikipedia tells me the T94 processors are worth 1.8gflops each, while the range-topping i7 is 109gflops.

Of course, in its day, intel and AMD were still in the middle of the mflops.

Funnily enough, there's a J932SE computer going for $16k - works, but crashes. 3.2 GFlops. http://www.ebay.com/itm/Cray-J932SE-32-processor-supercomput...

But will it run Crysis?

*Craysis

So cray.

That shit cray.

Reply on news.ycombinator.com