NNSA Supercomputer Breaks Computing Record

Lecutis writes "National Nuclear Security Administration (NNSA) Administrator Linton F. Brooks announced that on March 23, 2005, a supercomputer developed through the Advanced Simulation and Computing program for NNSAs Stockpile Stewardship efforts has performed 135.3 trillion floating point operations per second (teraFLOP/s) on the industry standard LINPACK benchmark, making it the fastest supercomputer in the world."

Neat

[neccoant]

It's amazing that we were stalled at 50TFLOPS for two years, and are piling on the FLOPS now.

Re:From the press release...

[Daniel Boisvert]

The closest I've heard of is the Cray X1E, but even that only claims [cray.com] 147 TFLOPS.

Re:hmmmmm...

[a1cypher]

Just for a point of reference, does anybody know how many floating point operations a 3.2ghz processor can do per seccond?

I know its not 3.2billion because most micro operations take at least 3 or 4 clock cycles.

Re:Neat

[woah]

The reason is, of course, that we've been stuck with sameish desktop performance as well. Which correlates with supercomputer performance, since nowdays most of them use Intel/AMD processors.

Just goes to show that Moore's law won't hold forever.

LINPACK usage?

[Gleepy]

I think of LAPACK [netlib.org] as being much more up-to-date for benchmarking.

Human Intelligence?

[kyle90]

Isn't the human brain supposed to be equivalent to a supercomputer running at about ~100 teraflops? And if so, shouldn't this computer be smarter than us?

Re:Human Intelligence?

[Kethinov]

Isn't the human brain supposed to be equivalent to a supercomputer running at about ~100 teraflops? And if so, shouldn't this computer be smarter than us?

In Star Trek TNG 2x09 [halo43.com] Data was quoted at having a total memory capacity of somewhere around 90 petabytes with a total linear computational speed of 60 trillian operations per second.

One would say this supercomputer is already more than twice as smart as Data!

Re:Wow!

[ucblockhead]

Heh. I guess I wasn't the only one who christianed a new machine by running fractint on it. Gave it up around 1998 because there was just no point.

Human Intelligence is More than Speed

[Ted Holmes]

One of the landmarks we needed to pass in order for computers to approximate Human intelligence is the processing speed.

Estimates are that the Human brain computes somewhere between 100 Teraflops and 1000 Teraflops [google.ca],
and Google was performing somewhere between 100 and 300 Teraflops [tnl.net]. in late 2004.

P.S. Since doing that bit of research, every time Google checks my spelling and responds with "did you mean..." the hair stands on the back of my neck :)

But it's more than processing speed. It needs to have the software to do things like decision making, analysis, reasoning, evaluating, judging, information-organizing, learning, logic etc. which would normally require a human to perform.

We're not far off though...

DOE's Senior Activity Center

[Animats]

The "stockpile stewardship program" is basically a senior activity center for retired physicists. They have busywork projects to keep people thinking about how to design nuclear weapons. DOE is worried that all the old bomb designers will die off, and no new ones will replace them.

Remember, everything in the inventory was designed with far less compute power than today's desktops.

NSA has something better

[Anonymous Coward]

When I worked at the NSA (I'm free to say this now because I live in Canada), I often heard the IT guys talking about how the supercomputer we used for sorting and decrypting telecommunications was faster than the ones they used at NASA, and a new cluster was in the planning stages to exceed 200 TFLOPS.

This was back in 2001.

I really have a strong feeling the NSA is still ahead of NASA on this one, but they don't publish information about their clusters... for obvious reasons.

Re:Neat

[imsabbel]

Well, in fact the truth is a right in the middle.
Linpack is VERY easy to parallize. Earth simulator and other vector machines get over 85% of their theoretical processing power with linpack, and even clusters with relatively abyssmal interconnects are still in the 50% range.

Lots of computational problems need orders of magnitutes more inter-node communication, up to the point where linpack doesnt even matter anymore and clusters and vector computers with the same linpack score are a factor of 10 or 20 apart.

Finally Beat SETI@Home

[billstewart]

Yes, I know that SETI@Home is actually running simpler problems than LINPAK, but it's still frustrating that the fastest computer in the world is being used for evil. Until this computer, the fastest computer in the world was a volunteer effort searching for space aliens using screen savers. Last November's numbers had the Blue Gene at 71 TFLOPS, which might have been faster than SETI (SETI's currently at 56 TFLOPS; I'm not sure if that's a decline or increase from last fall, given the number of people who have switched over to Folding@Home, various cancer-fighting applications, etc., vs. people who've gotten faster computers.) Number 2 was at NASA AMES, so it could have been working for good or evil applications, and Number 3 was the previous champion Earth Simulator, which is on the good side.

I'm currently running Folding@Home...