Beta
×

Welcome to the Slashdot Beta site -- learn more here. Use the link in the footer or click here to return to the Classic version of Slashdot.

Thank you!

Before you choose to head back to the Classic look of the site, we'd appreciate it if you share your thoughts on the Beta; your feedback is what drives our ongoing development.

Beta is different and we value you taking the time to try it out. Please take a look at the changes we've made in Beta and  learn more about it. Thanks for reading, and for making the site better!

NNSA Supercomputer Breaks Computing Record

timothy posted more than 9 years ago | from the we're-simulating-a-stockpile dept.

Supercomputing 266

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."

cancel ×

266 comments

Sorry! There are no comments related to the filter you selected.

Neat (3, Interesting)

neccoant (3345) | more than 9 years ago | (#12127013)

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

Re:Neat (3, Insightful)

the_2nd_coming (444906) | more than 9 years ago | (#12127036)

the increased flops is simply a function of the fact that they are expanding the number of nodes.

Re:Neat (3, Funny)

Anonymous Coward | more than 9 years ago | (#12127042)

I think they were waiting for final specs on Doom3.

Re:Neat (0)

Anonymous Coward | more than 9 years ago | (#12127065)

It's obviously to meet the heavy system requirements of the new MineSweeper bundled with Windows 64 bits.

Re:Neat (3, Interesting)

woah (781250) | more than 9 years ago | (#12127157)

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.

Re:Neat (0)

Anonymous Coward | more than 9 years ago | (#12127208)

Yeah, remeber that last time the terrorists had control of the override and Jack Bauer had to get it back. Without Jack and that fat bastard Edgar we would all be in deep shit now.

I guess after that happened they decided it was time for a computer upgrade.

Re:Neat (3, Insightful)

imsabbel (611519) | more than 9 years ago | (#12127228)

You are misstaken.
We didnt STALL at 30Gflops, its just that the 30Gflops were SO much better than everything else available that it took a couple of years to catch up and overtake it.

If you average over the last 10 years, the the Earth simulator was a bump above moores law and now we are back on track.

Re:Neat (0)

Anonymous Coward | more than 9 years ago | (#12127349)

We're back on track! We're back on track!

Huuuuzaaah!

Shit eating post (0, Troll)

Fecal Troll Matter (445929) | more than 9 years ago | (#12127016)

Don't forget to brush.

Re:Shit eating post (-1, Offtopic)

Anonymous Coward | more than 9 years ago | (#12127209)

FOSS sucks. Please get over it.

and its only half the machine too! (5, Informative)

rebelcool (247749) | more than 9 years ago | (#12127035)

wait till its fully online.

Re:and its only half the machine too! (1)

shadowsurfr1 (746027) | more than 9 years ago | (#12127164)

Holy shit. WTF are they going to use that kind of power for anyways?

Re:and its only half the machine too! (1)

shadowsurfr1 (746027) | more than 9 years ago | (#12127185)

Nevermind. Should've RTFA.

Re:and its only half the machine too! (2)

JPM NICK (660664) | more than 9 years ago | (#12127187)

Scientists at LLNL for the first time have performed 16-million-atom molecular dynamics simulations with the highest accuracy inter-atomic potentials necessary to resolve the key physical effects to successfully model pressure induced rapid resolidification in Tantalum. As a shared resource, BlueGene/L has enabled scientists studying the effects of voids in metal failure at Los Alamos National Laboratory to perform standard accuracy molecular dynamics simulations with more than 2.1 billion atoms.

Yes, but ... (-1, Offtopic)

ggvaidya (747058) | more than 9 years ago | (#12127038)

is that faster than a speeding bullet?

Sorry... (0, Insightful)

Anonymous Coward | more than 9 years ago | (#12127323)

Imagine a BeoWulf Cluster of these!!!

From the press release... (3, Informative)

Zebra_X (13249) | more than 9 years ago | (#12127040)

This performance was achieved at Lawrence Livermore National Laboratory (LLNL) at only the half-system point of the IBM BlueGene/L installation. Last November, just one-quarter of BlueGene/L topped the TOP500 List of the world's top supercomputers.

Is there anything that will be able to touch this when it's complete?

Re:From the press release... (4, Interesting)

Daniel Boisvert (143499) | more than 9 years ago | (#12127083)

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

Re:From the press release... (5, Informative)

As Seen On TV (857673) | more than 9 years ago | (#12127249)

The X1E isn't intended to be a fastest-in-the-world supercomputer. It's intended to be a low-cost scalable vector system. The fact that it's fast is great, but it's not its main design feature.

Now, the X2, on the other hand, is a whale. They're talking 150 TFLOPS at roll-out next year (unimpressive) and 300 TFLOPS after the block 10 update the year after that (very impressive).

Of course, the X2 isn't working yet, so who the hell knows. But it's fun to think about.

Re:From the press release... (1)

imsabbel (611519) | more than 9 years ago | (#12127383)

And thats in a hypothetical full version that nobody has ever build or bought...

Is there anything that will be able to touch this (0)

Anonymous Coward | more than 9 years ago | (#12127116)

"Is there anything that will be able to touch this when it's complete?"

RoboSapien may elicit some emotional response...

Re:From the press release... (1)

Dr. Weird (566938) | more than 9 years ago | (#12127134)

Is there anything that will be able to touch this when it's complete?

Probably the sysadmins...

Blue Gene? (2, Informative)

eth8686 (793978) | more than 9 years ago | (#12127045)

Didn't IBM push Blue Gene to 180'something teraflops recently?? News story herer [businessweek.com]

Re:Blue Gene? (0)

Anonymous Coward | more than 9 years ago | (#12127090)

Probably another benchmark/scientific program was run to achieve the different result.

Re:Blue Gene? (5, Informative)

EBorisch (530762) | more than 9 years ago | (#12127163)

This is Blue Gene. Read the article...

This *is* Blue gene. (2, Informative)

daveschroeder (516195) | more than 9 years ago | (#12127215)

RTFA [doe.gov]

Or, at least the article's title:

"NNSA Supercomputer Breaks Computing Record: Exceeds 100 TERAFLOPS DOE/NNSA and IBM partnership on BlueGene/L, a tool for national security"

Re:Blue Gene? (0)

Anonymous Coward | more than 9 years ago | (#12127285)

Didn't IBM push Blue Gene to 180'something teraflops recently??

Comments like this make slashdot suck. My comment on that comment sucking also makes slashdot suck. Thanks for taking initiative in making things worse around here.

AMazing (-1, Flamebait)

Anonymous Coward | more than 9 years ago | (#12127049)

Now we can make bigger and more efficient nuclear bombs. YEH!

Re:AMazing (1)

TheGavster (774657) | more than 9 years ago | (#12127112)

... or safer and more efficient nuclear reactors.

Re:AMazing (0)

Anonymous Coward | more than 9 years ago | (#12127177)

... you can ask the chinese for those.

Re:AMazing (0)

Anonymous Coward | more than 9 years ago | (#12127246)

or the iranians, they got it from the chinese.

Re:AMazing (2, Insightful)

mikael (484) | more than 9 years ago | (#12127473)

Or we can model protein folding even faster.
Run more accurate climate simulations even faster.
Run population simulations even faster.
Run CAD/CAM simulations even faster.

Wow (-1, Offtopic)

LiquidCoooled (634315) | more than 9 years ago | (#12127051)

Minesweeper has never looked so fast!

Or do nuclear scientists prefer solitaire?

imagine (3, Funny)

dario_moreno (263767) | more than 9 years ago | (#12127052)

a Beowulf cluster of these !

Re:imagine that (0)

Anonymous Coward | more than 9 years ago | (#12127070)

In Soviet Russia, supercomputers cool you.

Re:imagine (1)

OmgTEHMATRICKS (836103) | more than 9 years ago | (#12127127)

I, for one, welcome our new NASA overlords.

Re:imagine (0, Offtopic)

meza (414214) | more than 9 years ago | (#12127272)

1. Imagine a beowolf cluster
2. Welcome our new overlords
3. In Soviet Russia "3" is you
4. ...
5. Profit!

Re:imagine (0)

Anonymous Coward | more than 9 years ago | (#12127312)

In Korea, only old people use supercomputers.

Wow! (5, Funny)

FlyByPC (841016) | more than 9 years ago | (#12127054)

Just imagine running Fractint on this puppy!

Re:Wow! (0)

Anonymous Coward | more than 9 years ago | (#12127084)

I've never used Fractint, but my lowly 700 Mhz PC can render fractals in real-time with Xaos. [theory.org]

Re:Wow! (0)

Anonymous Coward | more than 9 years ago | (#12127392)

Fractals? I mean like, come on. Those are outfashioned now.

How about real-time global illumination using photon mapping instead? Now that would be impressive.

Re:Wow! (2, Interesting)

ucblockhead (63650) | more than 9 years ago | (#12127431)

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.

Steroids (4, Funny)

tiktok (147569) | more than 9 years ago | (#12127056)

There was another machine that had already beaten that record, but unfortunately failed a diagnostic test for banned substances...

Re:Steroids (1)

Yaotzin (827566) | more than 9 years ago | (#12127206)

Was the banned substances by any chance Skynet in liquid form?

Re:Steroids (0)

Anonymous Coward | more than 9 years ago | (#12127280)

You mean banned substrates...

Did you RTFA? (5, Informative)

Donny Smith (567043) | more than 9 years ago | (#12127058)

> 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."

Did you read the fucking article?

"This performance was achieved at Lawrence Livermore National Laboratory (LLNL) at only the half-system point of the IBM BlueGene/L installation. Last November, just one-quarter of BlueGene/L topped the TOP500 List of the world's top supercomputers."

See, this is the SAME supercomputer that has already topped the list last November, so the latest record did NOT make it the fastest supercomputer in the world.

It already had been the fastest supercomputer in the world.

Re:Did you RTFA? (1)

0x461FAB0BD7D2 (812236) | more than 9 years ago | (#12127162)

Did you RTFA yourself? The /. summary is pretty much the same as the first paragraph of the article.

Re:Did you RTFA? (1)

madprof (4723) | more than 9 years ago | (#12127240)

OK so it is still the fastest computer in the world.
Technically the description wsa accurate, however.

Re:Did you RTFA? (1)

Dun Malg (230075) | more than 9 years ago | (#12127375)

OK so it is still the fastest computer in the world. Technically the description wsa accurate, however.

Actually, technically the description was incorrect, as the term "making" requires that the subject initially not be what it was made into, i.e. not the fastest computer in the world.

But yeah, this is all splitting hairs, and I should be ashamed of myself for even mentioning this...

It's so fast... (1, Funny)

Anonymous Coward | more than 9 years ago | (#12127062)

that it almost meets the minimum requirements for Longhorn.

all the spppppppppeeed (-1, Flamebait)

Anonymous Coward | more than 9 years ago | (#12127067)

and I bet it just display % on the console :)

Wondering the only game on that box is the ol unix dungeon and dragon game.

Count On It (-1, Troll)

Doc Ruby (173196) | more than 9 years ago | (#12127077)

We'll need that performance to track the lies from the DoE about nuke waste insecurity [nytimes.com] . And the lies from the NRC about current nuke storage insecurity [washingtonpost.com] . Good thing our tax dollars are spent so wisely that we now have the situation under control.

Re:Count On It (-1, Offtopic)

Doc Ruby (173196) | more than 9 years ago | (#12127211)

Moderation -2
100% Troll

Trolls don't tell the truth, unlike my post. TrollMods don't like the truth, like my post.

Re:Count On It (0, Redundant)

As Seen On TV (857673) | more than 9 years ago | (#12127266)

Trolls post off-topic shit in an effort to stir up an argument. Just like your post.

hmmmmm... (0)

Rabid_Llama (873072) | more than 9 years ago | (#12127080)

kinda makes that 3.2 gh processor you thought was so fast piss its pants.

Re:hmmmmm... (5, Interesting)

a1cypher (619776) | more than 9 years ago | (#12127139)

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:hmmmmm... (3, Informative)

Yartrebo (690383) | more than 9 years ago | (#12127186)

With SSE instructions, you can process 4 floats at once, so I'm guessing that 3.2 GHz processor can do a few gigaflops.

Re:hmmmmm... (1)

Xoro (201854) | more than 9 years ago | (#12127238)

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

That used to be true, but with pipelining and parallelism you can no longer just divide the cycles per second by cycles per instruction and get instructions per second. Pipelining means that an operation does not have to finish before the next one is begun, and parallelism means that more than one operation can begin at the same time.

So now, the answer is to basically look it up.

Re:hmmmmm... (0)

Anonymous Coward | more than 9 years ago | (#12127335)

i remember reading something about the FP's on the P4 having no clock or atleast working ascycronusly

so your 3 4 cycles might be true but there is no way of knowing how fast they were done

Re:hmmmmm... (5, Informative)

tgamblin (873085) | more than 9 years ago | (#12127340)

Depends on the problem and the memory performance as much as it does on the GPU. There's no good answer to that question. For kicks though, this paper [stanford.edu] has some measurements for matrix multiply using ATLAS. It's comparing a Pentium 4 to an NV40 GPU. The P4 wins at about 7 GFlops, and the NV40 loses due to horrible memory performance. That's pretty ironic considering that the NV40 has quite a few more FPU's, and that they're in parallel. It's a good example of why you can't ever say for sure how a processor's going to perform until you test it on a real workload.

Re:hmmmmm... (2, Informative)

Swedentom (670978) | more than 9 years ago | (#12127377)

An Apple Xserve G5 does 30+ gigaflops.

Re:hmmmmm... (2, Informative)

fafalone (633739) | more than 9 years ago | (#12127478)

A 3.2GHz Intel Xeon processor performs 6.4gflops, but clock speed isn't the only determining factor.

Wow. (2, Funny)

TsukasaZero (850187) | more than 9 years ago | (#12127093)

Slap an X850 in there and you've got some serious Doom 3 action.

Re:Wow. (1)

jumbledInTheHead (837677) | more than 9 years ago | (#12127344)

Being that Doom isn't designed for parrallel processing it would probably run slower than your computer. Unless that was a joke, in which case I have no sense of humor.

Re:Wow. (1)

Professor_UNIX (867045) | more than 9 years ago | (#12127396)

Unless that was a joke, in which case I have no sense of humor.

Ding ding ding, we have a winnaahh. Pick a cupie doll for the little lady.

Re:Wow. (1)

Triffid_Hunter (746141) | more than 9 years ago | (#12127445)

at that rate of flops, i think software rendering would be orders of magnitude faster :P

heh heh (-1, Redundant)

Thaelon (250687) | more than 9 years ago | (#12127101)

Imagine a cluster of these!

doom3 (0)

Anonymous Coward | more than 9 years ago | (#12127122)

So am i still gonna have trouble playing Doom 3 on it?

Yeah, but... (3, Funny)

Black Jack Hyde (2374) | more than 9 years ago | (#12127129)

...will it run NetHack [nethack.org] ?

Re:Yeah, but... (1)

Fjornir (516960) | more than 9 years ago | (#12127173)

Why bother when you can play Angband [thangorodrim.net] ?

still not fast enough (2, Insightful)

ottawan (872094) | more than 9 years ago | (#12127147)

This computer is still way too slow for things like studying molecular structure, even with simplified models.

Earth Simulator (3, Insightful)

Anonymous Coward | more than 9 years ago | (#12127151)

I rather miss the time when the world's most
powerful supercomputer was used to study our
planet. It was something to be proud of, actually.
These machines are essentially weapons. Pity, that.

Re:Earth Simulator (2, Insightful)

lp-habu (734825) | more than 9 years ago | (#12127236)

Historically, I'll think you'll find that a great many technological advances were made with the original purpose of killing other beings -- usually other humans. Seems to be one of the basic human characteristics. Pretty effective, too.

Re:Earth Simulator (0)

Anonymous Coward | more than 9 years ago | (#12127398)

I know, and given that we're aware of this (as
opposed to unwittingly following instinct), doesn't
it make it even more tragic?

Re:Earth Simulator (1)

ramblin billy (856838) | more than 9 years ago | (#12127442)


Yeah, but you should have seen the shit the guys who got killed were working on.

It's been said that a modern handgun represents close to state of the art technology in many fields and costs about a weeks pay. The Icon for human progress?

billy - guns don't kill people - bullets do

Link to the list (4, Informative)

dnaboy (569188) | more than 9 years ago | (#12127160)

FYI the top 500 supercomputers list is maintained at http://www.top500.org/ [top500.org] .

More important issues (-1, Offtopic)

bogaboga (793279) | more than 9 years ago | (#12127161)

I wish our government and scientists dedicated more work to more important issues IMO, like pollution, healthcare, deficits and the injustices of the world, government perpetuated or corporate. As I write this, I figure I might be un-employed soon, as my job gets out-sourced to Russia. In America alone, almost 50 million of our residents do not have any kind of healthcare. With all these facts, we then go round the world "preaching" fairness and prosperity.

You are not seeing the larger picture (2, Insightful)

ravenspear (756059) | more than 9 years ago | (#12127197)

This is the same flawed logic that people try to apply to NASA on here a lot, that we shouldn't send people to the Moon/Mars until we fix all the problems on earth.

Halting scientific research to worry about all of our other problems is the wrong approach for many reasons. It is often scientific advances which lead to improved quality of life in many other areas of society.

Re:You are not seeing the larger picture (0)

Anonymous Coward | more than 9 years ago | (#12127216)

But then why not apply those supercomputing
capabilities to solve said problems...?

Re:You are not seeing the larger picture (1)

Dun Malg (230075) | more than 9 years ago | (#12127443)

But then why not apply those supercomputing capabilities to solve said problems...?

Because hunger is not a MATH problem.
Because civil wars are not a MATH problem.
Because disease is not a MATH problem.

There's only so much you can do with iterative models.

Re:You are not seeing the larger picture (1)

ramblin billy (856838) | more than 9 years ago | (#12127334)


What you say is true, but it's still kind of sad that we have to use the most powerful computer in the world to keep us from being contaminated by our own mess.

billy - wondering where that damn genie left the cork THIS time

Re:More important issues (1)

vhogemann (797994) | more than 9 years ago | (#12127247)

The irony is that all of that topics already have an solution... they're not put in pratice because of politics. Please, don't blame the scientists... blame the governament, or blame the people that elected the politicians that run your country!

So instead of writing this on Slashdot, why don't you make a telephone call to you Senator? Why don't you gather people that think like you and tell it to him?

If he got your vote, he might have a minute or two to listen what you has to say... You voted, didn't you?

Re:More important issues (5, Informative)

tgamblin (873085) | more than 9 years ago | (#12127295)

Despite the fact that BlueGene/L is being built to simulate nukes, this kind of research does impact some of these other issues, and there is government money going into them. Here are some examples... The National Center for Atmospheric Research uses supercomputers [ucar.edu] to simulate effects of pollution and global warming, and projects like LEAD [ou.edu] are using grids with supercomputers attached to predict weather. Check out some of the projects [renci.org] at RENCI, as well. There's NIH-sponsored genetic research in addition to the weather stuff.

It may be sad that we live in a world where nuclear weapons research is driving the computing power, but it doesn't mean that the power of BlueGene/L isn't going to be used for thousands of other peaceful scientific applications, too.

Re:More important issues (1)

bogaboga (793279) | more than 9 years ago | (#12127346)

The issue I have with this kind of reasoning is that even with the facts known, pollution especially goes on unabated! Our own president (Bush) chose to opt out of Kyoto. Meanwhile, some of our people are being found with diseases that were once unheard of just a few decades ago. Childhood diabetes is one, and what about juvenile cancer? The current research in fuel efficient vehicles and the like is driven not much by pollution, but by profits, when the price of oil reaches way up there. Who doesn't know this? Don't you?

Re:More important issues (1)

tgamblin (873085) | more than 9 years ago | (#12127370)

I never said that anyone would pay any attention to the research. You'll have to write your congressman for that. With the current administration in power I'm just happy that the research is getting done, so that maybe someone will notice it someday.

I think that as long as we have the research going on actively, there's still some hope. Once people stop caring about even finding out, then we're in trouble.

But.. (-1, Redundant)

Mad Ogre (564694) | more than 9 years ago | (#12127178)

But you still can't play GTA SA on it.

Re:But.. (0, Troll)

Anonymous Coward | more than 9 years ago | (#12127305)

who cares gta:sa sucks.

Dupe (3, Informative)

karvind (833059) | more than 9 years ago | (#12127199)

Didn't we cover this before [slashdot.org] ?

LINPACK usage? (2, Interesting)

Gleepy (16226) | more than 9 years ago | (#12127214)

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

Human Intelligence? (2, Interesting)

kyle90 (827345) | more than 9 years ago | (#12127254)

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? (1)

astro-g (548659) | more than 9 years ago | (#12127274)

Maybe,
but what working set does it have to work with?

Its not loaded with the right software, or the right data to think like a human, even if the hardware is capable of performing the right kind of operations, at any usefull speed.

Unsafe at any speed (0)

Anonymous Coward | more than 9 years ago | (#12127311)

Its not loaded with the right software, or the right data to think like a human, even if the hardware is capable of performing the right kind of operations, at any usefull speed.

Thinking machines...UNSAFE AT ANY SPEED!

Re:Human Intelligence? (1)

agent2 (628468) | more than 9 years ago | (#12127387)

Not necessarily. The speed of a system isn't completely related to the intelligence of a system. There are humans that think slower the other humans that are much more intelligent. So, speed helps, but the level of complexity isn't always related.

Re:Human Intelligence? (4, Interesting)

Kethinov (636034) | more than 9 years ago | (#12127404)

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!

But... (-1, Redundant)

Anonymous Coward | more than 9 years ago | (#12127281)

Yeah but can it run Linux? :)

"working to reduce global danger from ..." (0, Troll)

tyroneking (258793) | more than 9 years ago | (#12127435)

The NNSA "working to reduce global danger from weapons of mass destruction"
we'll be able to count all the WMDs in Iraq!
Hooray!

OK, let's get them all out of the way now (1)

Gudlyf (544445) | more than 9 years ago | (#12127449)

  • "Imagine a beowulf cluster of those!"
  • "Imagine the Seti@Home rank on that puppy!"
  • "Pfft... A mere abacus -- mention it not!
  • "Molest me not with this pocket calculator stuff!"

wake me up when we reach a petaflop (1)

peter303 (12292) | more than 9 years ago | (#12127455)

All these increment "my computer is faster than your computer" articles are getting boring. I'll be interested in when they reach a petaflop. With "Moore's law" predicting a 10x speed up every five years, that should be around 2010.
Load More Comments
Slashdot Login

Need an Account?

Forgot your password?

Submission Text Formatting Tips

We support a small subset of HTML, namely these tags:

  • b
  • i
  • p
  • br
  • a
  • ol
  • ul
  • li
  • dl
  • dt
  • dd
  • em
  • strong
  • tt
  • blockquote
  • div
  • quote
  • ecode

"ecode" can be used for code snippets, for example:

<ecode>    while(1) { do_something(); } </ecode>