Slashdot Mirror


Jaguar, World's Most Powerful Supercomputer

Protoclown writes "The National Center for Computational Sciences (NCCS), located at Oak Ridge National Labs (ORNL) in Tennessee, has upgraded the Jaguar supercomputer to 1.64-petaflops for use by scientists and engineers working in areas such as climate modeling, renewable energy, materials science, fusion and combustion. The current upgrade is the result of an addition of 200 cabinets of the Cray XT5 to the existing 84 cabinets of the XT4 Jaguar system. Jaguar is now the world's most powerful supercomputer available for open scientific research."

154 comments

  1. Silly Me by Anonymous Coward · · Score: 0

    I thought it was Leopard!

    1. Re:Silly Me by Jeremiah+Cornelius · · Score: 1

      No. Jaguar. 1995 XK12, Six-Litre.

      --
      "Flyin' in just a sweet place,
      Never been known to fail..."
    2. Re:Silly Me by colmore · · Score: 4, Funny

      Hahah, eat it, 3DO.

      --
      In Capitalist America, bank robs you!
    3. Re:Silly Me by damn_registrars · · Score: 2, Informative

      No. Jaguar. 1995 XK12, Six-Litre.

      I'd rather doubt the 1995 XK12, as cool as it was, was any competitor to the Jaguar XJ220.

      --
      Damn_registrars has no butt-hole. Damn_registrars has no use for a butt-hole.
    4. Re:Silly Me by Jeremiah+Cornelius · · Score: 1

      Yes! But the XJ isn't made of Unobtanium!

      --
      "Flyin' in just a sweet place,
      Never been known to fail..."
    5. Re:Silly Me by shnull · · Score: 0

      lol, and i thought i was the only one who remembered Atari :p

      --
      beware he who denies you access to information for in his mind, he already deems himself to be your master (SMAC-ish)
  2. Economics? by thedonger · · Score: 3, Interesting

    How about economic modeling?

    --
    Help fight poverty: Punch a poor person.
    1. Re:Economics? by Ethanol-fueled · · Score: 0, Redundant

      From the minimum-requirements-for-crysis dept.

      How about running Crysis...on VISTA!

    2. Re:Economics? by Anonymous Coward · · Score: 0

      Very nice, very topical, and almost completely impossible.

      Making an economic modeling system would be insane, and probably beyond even this machine's resources.
      Not to mention, an economic model would be a self fulfilling prophecy, with people selling of just before predicted downturns and thus creating them, even if they wouldn't happen without the "forecast."

    3. Re:Economics? by thedonger · · Score: 2, Insightful

      There already exists economic modeling. And it is no more impossible than climate modeling. Granted, human interaction becomes a factor when the general population is aware of the economic predictions, but I am talking theory, not necessarily practice.

      --
      Help fight poverty: Punch a poor person.
    4. Re:Economics? by Anonymous Coward · · Score: 0

      Obviously noone's been using supersomputers for economic modeling - how do you explain the mess the world is in?

    5. Re:Economics? by thedonger · · Score: 2, Funny

      how do you explain the mess the world is in?

      Pre-marital sex?

      --
      Help fight poverty: Punch a poor person.
    6. Re:Economics? by Anonymous Coward · · Score: 0

      The problem with those models is their lose connection to reality. They assume unlimited resources or everyone behaving rational.

    7. Re:Economics? by bhtooefr · · Score: 2, Insightful

      Accurate economic modeling needs infinite resources, as the existence of the economic modeler needs to be taken into account, and it could be argued that the entire universe would have to be modelled 100% accurately - one atom being in a different place could cause drastically different outcomes years down the line, causing different economic conditions.

    8. Re:Economics? by thedonger · · Score: 2, Informative

      Ultimately human behavior is near-continuous series of yes/no decisions. Our brains iterate pretty deeply, but at some level it's ones and zeros. Though we may need more petaflops than angels on the head of a pin before we can scratch that itch. At any rate, the application of such a model will probably always doom it to failure.

      How much do we really know about climate? Probably a lot less than we think. Scientists are always so sure they are right. And then a few decades pass and they realize they weren't. And then they repeat that same behavior.

      --
      Help fight poverty: Punch a poor person.
    9. Re:Economics? by mrsteveman1 · · Score: 2, Informative

      Every time you have sex outside marriage, god kills an economist (and a kitten)

    10. Re:Economics? by mr_death · · Score: 0

      Of course, the climate models have failed to correctly predict global temperate going forward. In particular, James Hansen's most famous prediction has failed miserably -- see http://climate-skeptic.typepad.com/photos/uncategorized/2008/09/25/hansen_forecast_1988.jpg , which calls into question Hansen's and Saint Gore's doomsday predictions.

      Curve fitting are easy; prediction is hard. Given the lack of prediction success of the climate models, true global economic models will probably have the same issues.

      --
      It's Linux, damnit! Pay no attention to renaming attempts by self-aggrandizing blowhards.
    11. Re:Economics? by treeves · · Score: 1

      I would suggest that marital sex results in offspring more frequently* than pre-marital sex, if you get my point.

      *more offspring per "event", on average.

      --
      ...the future crusty old bastards are already drinking the Kool-Aid.
    12. Re:Economics? by Anonymous Coward · · Score: 0

      Hey, that looks legit. Well, I'm convinced.

    13. Re:Economics? by Ambitwistor · · Score: 2, Interesting

      Look at the caption on your graph. Hansen's Scenario A is a high emissions scenario which does not correspond to the emissions which actually occurred. If you want to legitimately test the skill of a climate model, you need to compare apples to apples. In this case, Hansen's Scenario B is the one that most closely corresponded to the real emissions trajectory. (Since Hansen is a climate scientist, not an economist, he gave a range of possible emissions scenarios and did not claim the world would follow any specific one of them.) Even Steve McIntyre of Climate Audit acknowledges this.

      Your snide reference to "Saint Gore" indicates that your skepticism has more to do with your emotional biases than with any true scientific motivation. And citing a graph which makes a point of comparing a single month's temperature to another month's temperature makes me question your critical thinking skills. (Well, choosing to get your "science" from skeptic web sites instead of from the scientific literature is the main reason to question your criticial thinking skills.) But if you want to read some science, you could start here.

    14. Re:Economics? by Draek · · Score: 1

      The same could be said of any simulation of a system where both the system and the simulation itself aren't completely independant (and, one could say, always have been). But since that's an unrealistic expectation we approximate, and AFAIK in the case of economics we do it pretty well.

      --
      No problem is insoluble in all conceivable circumstances.
    15. Re:Economics? by Draek · · Score: 3, Interesting

      Scientists are always so sure they are right. And then a few decades pass and they realize they weren't. And then they repeat that same behavior.

      Not really. Most scientists know they're always wrong, they just try to be less wrong each time. Hence the scientific method.

      There's a brilliant article by Asimov about it, in fact, "The Relativity of Wrong" if you care about it.

      --
      No problem is insoluble in all conceivable circumstances.
    16. Re:Economics? by Anonymous Coward · · Score: 0

      I bet it won't boot in less than four seconds.

    17. Re:Economics? by Anonymous Coward · · Score: 0

      A systems view helps here. There is a difference in modelling an economic system constructed of laws and conventions to modelling each economic actor separately the same way there are differences of approach in modelling macro and micro economics.

    18. Re:Economics? by jacquesm · · Score: 1

      I would suggest that you can't be sure of that, especially not for firstborns. You'll have to do some math on an awful lot of people and when you do so you'll make quite a few enemies and find out that a large number of firstborns were conceived before their parents married, in fact were the reason their parents married.

      Try it!

      (and if you're a firstborn then you are the obvious test case ;) )

    19. Re:Economics? by RockWolf · · Score: 1
      I'm doing my best, here. Gimme a break, OK?

      Every day, a new economist pops up on tv...
      Less cats around, though. *grin*

      --
      February 9th, 2009 8:55pm: Slashdot becomes self-aware.
    20. Re:Economics? by eulernet · · Score: 1

      How about economic modeling?

      Nothing beats a pair of dice.

    21. Re:Economics? by KDR_11k · · Score: 1

      It's not certain that the mind is binary, it could very well be analogue.

      --
      Justice is the sheep getting arrested while an impartial judge declares the vote void.
    22. Re:Economics? by KDR_11k · · Score: 1

      But what do you have to do for God to kill a lawyer?

      --
      Justice is the sheep getting arrested while an impartial judge declares the vote void.
    23. Re:Economics? by mrsteveman1 · · Score: 1

      The fact that they are immune to gods wrath should tell you something about them....

    24. Re:Economics? by Anonymous Coward · · Score: 0

      The answer lies in pi and a power drill. Trust me.

    25. Re:Economics? by master_p · · Score: 1

      They tried, but the computer crashed.

    26. Re:Economics? by Mozk · · Score: 1

      And a 216-digit number? I suppose that worked out well for Max.

      --
      No existe.
    27. Re:Economics? by Jeremiah+Cornelius · · Score: 1

      How do you model PILLAGE and THEFT and CONFIDENCE GAMES?

      --
      "Flyin' in just a sweet place,
      Never been known to fail..."
  3. Yeah, the supercomputing stuff is nice and all... by Gizzmonic · · Score: 3, Insightful

    But I really got it to play Tempest 2000.

    --
    (-1, Raw and Uncut is the only way to read)
  4. How does that work? by TheDarkener · · Score: 1

    So, like, how do so many people use the computer effectively? Do they have a sign-in sheet? I bet there's a long line. :p~

    --
    It is pitch black. You are likely to be eaten by a grue.
    1. Re:How does that work? by Entropius · · Score: 4, Informative

      There is a queueing system. If you want to run a job on a machine like this, you log into the control node (which is just a linux box) and submit your job to the queue, including how many CPU's you need for it and how much time you need on them.

      A scheduling algorithm then determines when the various jobs waiting in the queue get to run, and sends mail to their owners when they start and stop.

      On many machines there is a debug queue with low limits for number of CPU's and runtime, and thus fast turnover; this is used to run little jobs to ensure everything is working right before you submit the big job to the main queue.

      Each project has an al

    2. Re:How does that work? by TheDarkener · · Score: 1

      Wow, I was actually joking, but ... cool. I actually thought it worked more like a mainframe/*nix terminal server.

      --
      It is pitch black. You are likely to be eaten by a grue.
    3. Re:How does that work? by CodeBuster · · Score: 1

      Do you need special permission to request ALL of the processors for your job?

    4. Re:How does that work? by Anonymous Coward · · Score: 1, Informative

      NCCS is a capability site, so no. You just need to be willing to wait for your job to bubble up to the top of the queue. In fact, as a capability site, the whole point is to develop codes that can run on the entire machine. Now, once your job runs, you will have to wait a while to get another opportunity, as the queues are set up to provide an 8 week moving average of "fairness".

    5. Re:How does that work? by Anonymous Coward · · Score: 0

      yes. usually, because otherwise you'd never get them all. These are high utilization systems, so there are generally lots of jobs running.

      For big, capability-class runs (ones that use the whole machine) you usually need to first prove that your application has a chance of scaling to a system that large, then get special permission to run it. It might also depend on what project you're on; i.e. does your application's function line up with NCCS's goals?

      NCCS is funded by office of science, so that includes most open science code. You would stand less of a chance of running something big on BlueGene at Lawrence Livermore or Roadrunner at Los Alamos.

  5. yes but does it run vista/crysis by wjh31 · · Score: 1

    probably not, its not designed for that sort of thing so its a stupid question to ask, nd probably uses linux anyway

    well that got that one out the way

    1. Re:yes but does it run vista/crysis by Entropius · · Score: 1

      It can't, because Crysis is not multithreaded. If you can figure out a way to parallelize it, then you certainly could run Crysis on it.

    2. Re:yes but does it run vista/crysis by Hurricane78 · · Score: 1

      Hell, you could *simulate* an x86_64 CPU on that thing, and it would not even hiccup on Crysis on Vista with every setting set to "Extreme".
      I wonder if someone could replace the engine by a full global illumination raytracer first. :D

      --
      Any sufficiently advanced intelligence is indistinguishable from stupidity.
    3. Re:yes but does it run vista/crysis by Anonymous Coward · · Score: 0

      why simulate ? look up the tech specs from TFA, the thing is actually rigged with thousands of AMD Opteron processors

    4. Re:yes but does it run vista/crysis by saratchandra · · Score: 1

      The login nodes run Linux but the compute nodes run CNL (Compute Node Linux), a lightweight OS designed to reduce system overhead. On an earlier version of this machine (XT3), the compute nodes used to run Catamount, a different lightweight OS.

    5. Re:yes but does it run vista/crysis by jacquesm · · Score: 1

      I don't think you have a clue how such a machine works. To simulate a non-trivially-parallelizable system like an x86_64 on it you could run your simulation on exactly the same number of cpu's as you have cores in your simulated cpu, or less (for a much easier to program simulation).

      Clusters such as these are *built* using standard hardware, in this case a (rather large) bunch of AMD opterons.

      If you can show me how to run 'Crysis on Vista' faster on rack with 10 pcs without rewriting either (and I did not see a cluster version of either) then your claim stands, otherwise it just confuses the issue of what a super computer is and what it can do.

      A typical task for a computer such as this one has the main task broken up into a large number of subtasks so that each CPU in the cluster can work on a small portion of the problem at the same time. This results in a speedup of some fraction of the total number of cpus versus the same time to run the problem on a single cpu. To achieve any meaningful speedup you have to be able to break the problem up like this, otherwise all that shiny hardware will just end up waiting for each other.

  6. Don't buy it by CRCulver · · Score: 3, Funny

    Yeah, Jaguar might look cool with its advanced capabilities, but there's no games for it and the controller design is lame.

    1. Re:Don't buy it by Z80a · · Score: 1

      and the hardware itself its quite buggy,as you need to align the jumps with 64 bits words but at least there is tempest for it

    2. Re:Don't buy it by LifesABeach · · Score: 1

      So, I'm standing in front of my new Jag, with my WOW CD in my trembling hands. Where do I plug in my Game Keyboard, and Mouse? There's nothing in the owners manual about where the plug is to connect to my Cable Box!? How much was this thing again?

      And another thing, where is the Cup Holder?

    3. Re:Don't buy it by AKAImBatman · · Score: 1

      And another thing, where is the Cup Holder?

      It was eliminated from this model. But you can buy an extra toilet seat!

    4. Re:Don't buy it by Anonymous Coward · · Score: 0

      Were current and former employees who worked on the ATARI JAGUAR 64-bit gaming system involved in this supercomputer project, hence the name Jaguar?

    5. Re:Don't buy it by Anonymous Coward · · Score: 0

      So, is it a true 64bit system this time?

    6. Re:Don't buy it by keeboo · · Score: 1

      Yeah, the CD-ROM accessory was joked to exhaustion back then.

  7. Please no climate modelling! by backslashdot · · Score: 0

    Why do climate modelling?

    We know global warming happens, we know how to reduce it. WTF. There is already a wealth of evidence for global warming, whats lacking is political will .. which wasting resources on climate modelling won't bring about .. simple education will.

    Bigger payoffs are drug discovery and fusion energy research.

    1. Re:Please no climate modelling! by Arthur+B. · · Score: 2, Insightful

      There is already a wealth of political will for global warming, whats lacking is evidence ..

      There, fixed that for you.

      --
      \u262D = \u5350
    2. Re:Please no climate modelling! by Anonymous Coward · · Score: 0

      Too bad you weren't around for the last ice age or you could have led the charge to warm up the earth.

    3. Re:Please no climate modelling! by thedonger · · Score: 3, Insightful

      I would say more than lack of evidence is lack of causation rather than correlation. Scientists appear to agree that at least in the short term the earth is a little warmer. What they can't say with any certainty is why. Anthropogenic warming is the desired cause as that is the only one we can do a damn thing about.

      --
      Help fight poverty: Punch a poor person.
    4. Re:Please no climate modelling! by Darth · · Score: 2, Insightful

      Who says the climate modeling they are doing is related to global warming?

      Even if it is, however, if the modeling increases our knowledge of the subject, it is not a waste of resources for scientists to seek the answers they are looking for.

      --
      Darth --
      Nil Mortifi, Sine Lucre
    5. Re:Please no climate modelling! by leathered · · Score: 4, Funny

      Why do climate modelling?

      Obviously climate modelling has to be carried to out to find out what impact running energy-hungry supercomputers has on the environment.

      --
      For all intensive porpoises your a bunch of rediculous loosers
    6. Re:Please no climate modelling! by reiisi · · Score: 1

      And the only cause that is politically useful.

      --
      Computer memory is just fancy paper, CPUs just fancy pens with fancy erasers; the 'net is just a fancy backyard fence.
    7. Re:Please no climate modelling! by Bunny+Guy · · Score: 2, Informative

      Having done my graduate work in fluid dynamics, only half your statement is possibly correct. There is historical evidence for global climate change, both warming and cooling. If is our interest to maintain the current status quo, the climate as we know it, it is not at all clear what interventions we need to take, or what effect they might have. Without simulations, that correctly model the real world. We have no way of knowing what our interventions might do. If anyone is interested, I can elaborate. The short (and scary)answer is - resonance.

    8. Re:Please no climate modelling! by Anonymous Coward · · Score: 0

      "There is already a wealth of political will for global warming, whats lacking is evidence..there, fixed that for you"

      Bullshit. Sorry, but you need to get your facts right before stuffing your head up your ass. There is hardly any political will to go ahead with research and development and (key word) implementation of novel technologies that could help us and the environment. Why? Money.

    9. Re:Please no climate modelling! by a_claudiu · · Score: 1

      Anthropogenic warming is the desired cause as that is the only one we can do a damn thing about.

      People can do something even if the warming is not man made. If the cause is proven to be man made maybe the people will be persuaded to react faster.

    10. Re:Please no climate modelling! by LingNoi · · Score: 1

      All good points but lets not forget that simulation != observation.

      This point was made about astrological observations on older slashdot stories and it holds true for climate prediction too.

      I have to admit I am sceptical about blindly believing in global warming. I used to in the past however I've become a little smarter since then I can not see any hard observations for it, especially when volcanoes pump out 26 times more CO2 per year then all of humanity on the planet (however I'm slightly sceptical of how they measure the pollution figures too).

      Wait, what's this got to do with the story... : /

    11. Re:Please no climate modelling! by Ambitwistor · · Score: 1

      Why do climate modelling?

      Well, purely scientific reasons for one. Climate science existed long before global warming was a concern.

      Another reason is to inform adaptation. No matter what policy is realistically put in place, at least some more climate change is expected to occur. People are going to have to adapt to whatever change is not prevented. It's thus important to improve our understanding of what may happen. It also tells us how large and how fast a policy response is required, although as you note, we are not yet even attempting the minimum recommended policy.

    12. Re:Please no climate modelling! by Ambitwistor · · Score: 1

      Global warming is not based merely on correlation studies. It has a direct and well understood physical cause, which is the greenhouse effect. (What is less understood is the climate system feedbacks which modify the greenhouse effect.) And climate scientists can indeed say with a high degree of confidence that the recent warming is due mostly to human activities. This evidence comes from physical reasoning as well as observational measurements (such as the stratospheric cooling signature of the enhanced greenhouse effect), as well as a quantification of manmade and natural sources of warming and cooling.

    13. Re:Please no climate modelling! by Ambitwistor · · Score: 1

      I have to admit I am sceptical about blindly believing in global warming. I used to in the past however I've become a little smarter since then I can not see any hard observations for it, especially when volcanoes pump out 26 times more CO2 per year then all of humanity on the planet

      That's not even remotely true. Volcanic CO2 emissions are about 1% of human CO2 emissions (see here). Where did you get the rather specific, and wrong, factor of 26?

      Maybe you'd see the hard evidence if you spend a little more time reading about it, since you appear to have some peculiar misconceptions. I recommend Kerry Emanuel's essay "Phaeton's Reins", David Archer's undergrad textbook, and the IPCC AR4 report for technical details.

    14. Re:Please no climate modelling! by Ambitwistor · · Score: 1

      Who says the climate modeling they are doing is related to global warming?

      Oak Ridge does: "Climate scientists are calculating the potential consequences of greenhouse gas emissions and the potential benefits of limiting these emissions."

    15. Re:Please no climate modelling! by Arthur+B. · · Score: 1

      To clarify myself, I do not have a particular stance on global warming itself. I do not know climatology. I do believe however that

      a) Appeal to consensus are irrelevant, scientific truth comes from repeated and verified experiments, not consensus.

      b) Whether it is man made or not is an ethical puppet to distract attention. What's done done.

      c) Most of the politics and economics surrounding global warming are flawed. That much I can tell. Even the worst prediction of the IPCC are quite tame compared to other disasters likely to happen, and they are dwarfed by the difference that even 0.5% of growth and technological progress a year makes on the same timescale.

      Most of the global warming debate is really an attack on industrialization and global trade under the guise of environmentalism.
      (And no it doesn't mean there's no global warming, and no it doesn't mean we shouldn't do something about it, but so far the debate has been mostly a trojan horse for political ideology).

      --
      \u262D = \u5350
    16. Re:Please no climate modelling! by Ambitwistor · · Score: 1

      The worst IPCC scenario (A1FI) gives a worst case of 6.4 C (11.5 F) global warming in less than 100 years. I don't know if it's the worst thing that is likely to happen, but that's not that tame (especially when you consider that land warms faster than the global average, and northern latitudes even faster than that). As for the economics, Nordhaus's book A Question of Balance is a good place to start. Nordhaus isn't ideological. It's economically worth mitigating some CO2 emissions to insure against the more severe outcomes. Not cutting them to zero instantly, but reducing them. Even the so-called Copenhagen Consensus, which concluded that a marginal investment in other disasters is more cost effective than pure CO2 abatement for global warming, ended up recommending CO2 abatement (in combination with adaptation and tech R&D).

      And frankly, I don't particularly care whether certain parties find global warming ideologically convenient. The problem still exists.

    17. Re:Please no climate modelling! by Nursie · · Score: 1

      Whether it's man made or not is quite important, actually. If it's not (a scenario that's looking pretty damned unlikely) then doing something to halt or slow it becomes difficult as we have to find out what the hell IS causing the problem.

  8. Non-Obligatory Frisky Dingo Reference by MaxwellEdison · · Score: 1

    Go Team Jaguar!

    Boosh!

    --
    -=Bang Bang=-
  9. Kudos to Atari by rgo · · Score: 5, Funny

    I always knew you could pull it off!!

    1. Re:Kudos to Atari by ari+wins · · Score: 2, Funny

      Dude, Altered Beast on this thing is SICK!

      --
      Don't worry if you're a kleptomaniac, you can always take something for it.
    2. Re:Kudos to Atari by joocemann · · Score: 0

      I always knew you could pull it off!!

      That's nothing. I'll have the world's most powerful supercomputer in a month. This is what I do... See.. they took 200 cabinets and added that to the already present 84....

      This is how I win. I add 400 cabinets!

      AHAHAHHAHAa!!! I GOT THE MOST POWERFUL ONE NOW! MUAHAHAHAHAH 200 ain't shit on my 400! ...... I just hope noone else can imagine a number greater than 400 and beat me....

    3. Re:Kudos to Atari by Anonymous Coward · · Score: 0

      Altered Beast is a Sega game, and ran on the Master System and Genesis/Mega Drive.

  10. Well done Atari by Anonymous Coward · · Score: 0

    Some people doubted the Jaguar, but I kept the faith. It's 64 bit - do the math on that!

  11. good upgrade path by Jodka · · Score: 2, Insightful

    The current upgrade is the result of an addition of 200 cabinets of the Cray XT5 to the existing 84 cabinets of the XT4 Jaguar system.

    That sounds like Cray engineered this to aggregate components across product generations. For short product life cycles that seems like a great idea, not throwing out the old system when you get the new one but combining the two systems instead. Though obviously for long product life cycles it would be a losing proposition; The space and power requirements of inefficient older components would be greater than the space and power savings of upgrading to the latest model + the expense of the upgrade.

    --
    Ceci n'est pas une signature.
    1. Re:good upgrade path by bmwm3nut · · Score: 3, Insightful

      Then when you get new cabinets you just decommission the oldest ones. Keep rotating the old ones out once they fall below some flop/dollar threshold.

  12. translation???? by pablo_max · · Score: 1

    Can someone please translate the performance into people/hand held calculator/time and space into number of libraries of congress? I am not sure what the numbers they're talking about mean.
    Thanks.

    1. Re:translation???? by The+Master+Control+P · · Score: 2, Informative

      Assume every single person on earth can do a 16-digit operation on a calculator in one second.

      It would take them roughly a quarter million years with no breaks of any kind to do what this machine can do in one second.

    2. Re:translation???? by The+Master+Control+P · · Score: 4, Funny

      Wait, shit, astronomical fail!

      It's not nearly that bad... more like 3 days. I failed to realize that my 270000 figure was seconds not years.

    3. Re:translation???? by philipmather · · Score: 0

      No, no it's EPIC FAIL! Pronounced in the same way the voice over from Mortal Kombat said "FINISH HIM!"

      --
      Regards, Phil
    4. Re:translation???? by elgatozorbas · · Score: 1

      This is the reason why we let the supercomputer do the calculations...

  13. I knew.. by Tmack · · Score: 1
    Atari would make a comeback!!

    Tm

    --
    Support TBI Research: http://www.raisinhope.org
  14. !scimanydomreht fo wal dnoceS by Nux'd · · Score: 1

    Will environmentalists ever stop trying to reverse the second law of thermodynamics?

    "Insufficient data for a meaningful answer."

    Damn.

  15. Re:Prediction: I will pound pussy tonight by Anonymous Coward · · Score: 0

    I hear that she prefer's it in the ass.

  16. "Used for open science..." by Entropius · · Score: 0

    There are much bigger computers around in Los Alamos and Lawrence Livermore that are used to model thermonuclear weapons sitting around in storage to see if they'll still go pop when we push the Big Red Button.

    The scientific community would like to use these machines for something useful, and in fact the scientists at Los Alamos have allowed some folks from my lattice QCD group to use a bit of spare time on it. Unfortunately the UNIX security features aren't enough; they weren't allowed to ftp our data out, and instead had to send the guy with the security clearance in with a pencil and paper to write down a page full of numbers. This is because of the ludicrous security procedures surrounding this Classified Machine. As you might imagine, this isn't a terribly efficient way to do things in a field where tens of gigabytes are sftp'd across the country without blinking.

    If we are ever in a situation where we're genuinely concerned with whether or not a given one of our nukes will still go pop, we are Already Fucked (tm). That's like modelling exactly what happens if a 10-km-radius asteroid hits the earth: at that point, we don't really care.

    This isn't to say that the nation's resources should be used for building giant computers for scientific research. The discussion about how to allocate our research funding is a difficult one. But if we're going to invest so much in computational power, it seems prudent to use it for something other than checking to see whether our e-peen is still hard.

    It says something depressing about our country that the top of the supercomputer list is used predominantly for military purposes.

    1. Re:"Used for open science..." by momerath2003 · · Score: 2, Informative

      You, sir, are an idiot.

      LANL, LLNL, and SNL are all weapons labs. ORNL is primarily a science lab.

      I myself have worked at three of these labs and held an account on an earlier iteration of Jaguar as well as some of LANL's other supercomputing clusters, so I ought to know.

      ORNL's Jaguar cluster, although parts of it are I think "controlled" rather than open so that it can run export-controlled code, is not at all classified. It's used for biology, astronomy, physics, CFD, etc.

      Also, if you knew the first thing about classified security you would realize that disallowing FTP access on a *classified* (Red network) machine to the outside internet is a necessity. To my knowledge, they don't allow *any* interconnection between classified systems and unclassified.

      --
      I had but a simple dream, to destroy all humans.
    2. Re:"Used for open science..." by caerwyn · · Score: 1

      You, sir, can't read.

      He was specifically talking about LANL and LLNL rather than ORNL.. that was the entire point.

      Granted, yes, his description of disallowing classified non-classified connectivity as "ludicrous" is a little off-base, although writing things down by hand really is stupid- there are plenty of procedures in place for putting data on transportable media and then arranging to declassify that media once it has been verified so that it can be used elsewhere.

      That doesn't change his fundamental points, though, that there are larger computing clusters that are locked up in classified nuclear simulation that could (and perhaps should) be used for more general science.

      --
      The ringing of the division bell has begun... -PF
    3. Re:"Used for open science..." by momerath2003 · · Score: 1

      In that case, pardon my misunderstanding in thinking that his post was at all related to the posted article.

      His title, "Used for open science...", was a quotation from the summary specifically about the ORNL computer. His rant about "much bigger computers around" was plausibly interpreted as the biggest one, the new ORNL cluster. I certainly must have been misled.

      --
      I had but a simple dream, to destroy all humans.
    4. Re:"Used for open science..." by jacquesm · · Score: 1

      So, let me get this straight: If someone writes something you think is not true he or she is an 'idiot', whereas if you write something that you admit wasn't true you 'have been misled' ?

  17. Re:Prediction: I will pound pussy tonight by Anonymous Coward · · Score: 0

    He said girlfriend, not mother.

  18. It might not look like much... by Anonymous Coward · · Score: 0

    ...but it can find Sarah Connor in under 6 parsecs.

    1. Re:It might not look like much... by Faylone · · Score: 1

      So, it can find her in under four times the distance to Alpha Centauri? Way to aim for precision.

    2. Re:It might not look like much... by keeboo · · Score: 1

      It depends on the context.
      If you are searching the whole Local Group, it seems reasonable.

  19. Slashdotted by Anonymous Coward · · Score: 0

    Meh. The most powerful supercomputer and the site still takes forever to open.

  20. Upgrade! by Chris+L.+Mason · · Score: 1

    They should really upgrade, Jaguar is ancient!

  21. 64-bit Jaguar by Anonymous Coward · · Score: 1, Informative

    remember that its not a true 64-bit multimedia system its two 32-bit systems connected together XD

  22. Re:Prediction: I will pound pussy tonight by Anonymous Coward · · Score: 0

    po-tay-to po-tah-to?

  23. So cool! by darthwader · · Score: 0, Redundant

    Imagine a Beowulf cluster of these!

    --
    I hate it when I make a joke and I get modded "+5 insightful". Mod the stupid comments "funny", not "insightful", pleas
    1. Re:So cool! by Arthur+Grumbine · · Score: 1

      I am so confused by the mods...the post above this one (posted 1 minute before) says exactly the same thing but gets "-1 Redundant"?!! Is it just because it was an AC, or did the title of "So cool!" really change the joke?!?

      --
      Now that I think about it, I'm pretty sure everything I just said is completely wrong.
    2. Re:So cool! by xonar · · Score: 1

      Maybe they thought someone who could actually receive karma deserved it more?

    3. Re:So cool! by Anonymous Coward · · Score: 0

      If I still had mod points I'd mod him redundant, fwiw.

  24. Folding@Home Contribution? by CodeBuster · · Score: 2, Insightful

    It would be nice on these sorts of systems to have recurring, perhaps low priority, jobs issued by worthy outside distributed computing projects. Depending upon how busy the system is with other jobs it could make regular contributions to drug research and especially to AIDS research. To have complete and accurate pre-computed models of all steps in the protein folding process for all possible mutations of the AIDS virus, for example, would be a technological triumph and of potentially great benefit to humanity in the development of new drugs and possibly even an effective vaccine.

    1. Re:Folding@Home Contribution? by Adam+Hazzlebank · · Score: 1

      To have complete and accurate pre-computed models of all steps in the protein folding process for all possible mutations of the AIDS virus

      1. Each trajectory would be several terabytes (possibly verging on petabypes).

      2. The largest simulation I know of is this one: http://www.ks.uiuc.edu/Research/STMV/ they simulated for 50ns and it's 10 times smaller than HIV. Protein folding takes milliseconds, not nanoseconds... it's not really tractable right now. I don't know how much cpu time the simulation took but it would have been a lot.

      3. Clusters like these are rarely idle, jobs are queued up to run when the cpus become available.

    2. Re:Folding@Home Contribution? by flaming-opus · · Score: 1

      There's no reason, whatsoever, to use a highly-connected, high-bandwidth HPC machine, like Jaguar, on distributed computing jobs. There are other very worthy jobs that can be run on such a system, that can't be run on a pile of desktops all over the internet. Use the real supercomputers for real supercomputer jobs. There are plenty of idle xbox in the world for distributed computing.

    3. Re:Folding@Home Contribution? by Anonymous Coward · · Score: 0

      They do. NCCS isn't on it, but the TeraGrid has a number of very large supercomputers you can submit jobs to from most anywhere. The largest of these is Texas's Ranger system (500-600 Teraflops). See here for a list of resources:

      http://www.teragrid.org/userinfo/hardware/resources.php

  25. what, no 'Vista capable' jokes? by catmistake · · Score: 0

    And I downloaded the summary and all the comments too! wtf?

    1. Re:what, no 'Vista capable' jokes? by badboy_tw2002 · · Score: 1

      Awww shit dude, your day is totally ruined! Don't worry, I think Crysis and Beowulf made it in. Phew! I thought /. was losing its edge!

  26. Hmm... by Troll14 · · Score: 0

    "1.64-petaflops for use by scientists and engineers working in areas such as climate modeling, renewable energy, materials science, fusion and combustion." Oh comon, do you think that there going to use ALL of this storage for science? I bet there's a few hundred TB worth of pron on there...

    --
    "Mama always said life was like a box a chocolates, never know what you're gonna get" - Forest Gump
  27. Love the paint job! by neowolf · · Score: 2, Informative

    Check out the gallery if you haven't.

    I've always wanted to get some custom graphics like that on my server racks. Maybe a penguin, a butterfly, and a can of Raid. :)

    Supercomputers definitely don't look as exciting as they did in the "old days".

    1. Re:Love the paint job! by cbreaker · · Score: 1

      Yea man, that thing is fucking awesome. I love the graphics.

      Who said super computing had to be boring?

      That's the kind of work I wish I could be doing.. building and configuring those monsters.

      That being said, look at the size of that room! If half of it is the computer, then that's one big open space left doing nothing.

      I'm sure they'll use it for something eventually.

      --
      - It's not the Macs I hate. It's Digg users. -
  28. not that great by LingNoi · · Score: 0

    It's been offline a lot this month...

  29. Just one thing by damaki · · Score: 1

    42

    --
    Stupidity is the root of all evil.
  30. 181,504 Opteron cores! by 2ms · · Score: 1

    I wonder how much they paid for all those Opterons. I wonder what kind of volume discount is typical for these kinds of supercomputers.

    1. Re:181,504 Opteron cores! by jacquesm · · Score: 1

      The real trickery is in the interconnects and the cooling, the cpu's are probably discounted quite a bit but I doubt they're the biggest item on the tab.

  31. Climate modeling ves. fusion energy by backslashdot · · Score: 1

    If we can get fusion energy working cheap, we won't need the climate modeling. Not only that we can build a hundred of these things cheaper with the technology advances.

    Climate change is gradual .. the need for new drugs and fusion energy is more pressing.

    1. Re:Climate modeling ves. fusion energy by Ambitwistor · · Score: 2, Interesting

      Climate change is gradual, but the emissions we put into the atmosphere today will last for centuries. Even if we switched over to all fusion power tomorrow, we'd still see more climate change, and the longer we wait to replace fossil fuels, the more we will see. Realistically, it takes a long time to widely deploy a new energy technology. Fusion isn't even feasible in the lab, let alone ready for deployment, let alone widely deployed.

      Also, even if fusion were widely deployed, that doesn't mean we'd necessarily have less fossil fuel emissions. Coal plants are cheap because they're already built, so we might just keep running them instead of shutting them down and having to build a new fusion plant, even a cheap one. They typically have operating lifetimes of over 50 years.

  32. Not that hard... by Junta · · Score: 1

    These systems are not so tightly integrated as you may imagine. True, many size a full-speed fabric just-right, each little bit costs a ton. However, commonly at scale, you only have full-speed fabric in large subsections anyway, and oversubscribe between the subsections. Jobs tend to be scheduled within subsections as they fit, though the subsection interconnects are no slouch.

    This is particularly popular as the authortitative Top500 benchmark is not too badly impacted by such a network topology, and real life workloads tend to not be so large as not to fit in a subsection (just all the subsections would be independently at work).

    It's kind of akin to saying you designed a couple of desktops well, because you could 'expand' them to host a lan party by hooking more to switches. Not quite so trivial, but the supercomputers aren't that especially exotic either.

    --
    XML is like violence. If it doesn't solve the problem, use more.
    1. Re:Not that hard... by convolvatron · · Score: 1

      i can easily imagine the hardware people making the interconnect lanes be backward compatible, but i believe there has been some endpoint capability changes between the two generations.

      whats difficult to believe is that they actually managed to encorporate the necessary changes in the resource allocator

  33. What's the score? by Junta · · Score: 1

    I noticed this a short time ago, but have yet to see the 'Rmax' performance. They speak to Rpeak, which does beat out the current Rpeak by 23%, though the Rpeak by itself is even more uninformative than Rmax, which is already quite synthetic. Assuming the current #1 hasn't managed tuning or upgrades, this will have to beat 65% efficiency to technically win. 65% is likely an acheiveable goal, though the larger the run, the more difficult to extract a reasonable efficiency number, so it's not certain. I wouldn't expect them to be so loud about it unless they knew the score already though...

    I will be interested to see the power consumption figures if offered. For that 25% increase in flops, they are requiring well over twice as many processor packages than RoadRunner (about 18,000 sockets vs 45,000 sockets here).

    I do wonder if Cray will be migrating to Intel in the next year given the QPI situation. AMD hasn't kept up compute leadership, and now HT will be lagging performance wise QPI.

    --
    XML is like violence. If it doesn't solve the problem, use more.
  34. Time to act like a user by rhenley · · Score: 1

    Hmmm...the ORNL web site lists the phone number for the Help Line. I think someone should call them up and ask them to reboot the server because the Internet is running slow.

  35. SETI by madcat2c · · Score: 2, Funny

    That is a lot of SETI@home power!

  36. As for me... by dark42 · · Score: 1

    I think I'll wait for Leopard.

  37. AMD vs Intel by Jimbookis · · Score: 1

    Someone care to explain why this uses Opterons instead of something Xeony? What does an AMD chip do for supercomputers that it can't seem to do for games and desktop machines?

    1. Re:AMD vs Intel by kpesler · · Score: 3, Informative

      I believe Cray made its partnership with AMD quite a while ago while they were still ahead of Intel in the performance/power ratio. In addition, these machines have a very fast interconnect (SeaStar) that is based on HyperTransport links. I believe it was recently announced that Cray has formed a partnership with Intel, and I imagine they will port the technology to QuickPath for future machines, but QPI was not available at the time this machine was commissioned. One does not simply order a machine like this at the drop of a hat. Vendor decisions are typically made years ahead of time.

  38. Should be good as long as Lucas wasn't involved by ObitMan · · Score: 1

    Lucas Electronics are crap for Jaguar and MG electrical systems. /Jaguar XJ owner

    --
    Who run Barter Town?
    1. Re:Should be good as long as Lucas wasn't involved by zzatz · · Score: 1

      Why do Brits drink ale at room temperature?

      They have Lucas refrigerators.

  39. I bet the guys at... by glitch23 · · Score: 2, Informative

    Los Alamos are jealous since they just got a 1.026 petaflop supercomputer installed earlier this year by IBM called RoadRunner. It was featured in last month's Linux Journal.

    --
    this nation, under God, shall have a new birth of freedom. -- Lincoln, Gettysburg Address
  40. Avoid the ground by superstition222 · · Score: 1

    Ouch

  41. So, what's the operating system? by dsmall · · Score: 1

    I hate to even ask, but ...

    What operating system does this thing run?

    I find myself hoping that's it's Linux with a Beowulf cluster, but ...

      -- thanks,

    David Small

    1. Re:So, what's the operating system? by DanZ23 · · Score: 1

      SuSE Linux

  42. How do we generate power? by Anonymous Coward · · Score: 0

    We burn stuff and [generally] use that HEAT to make steam to turn turbines, push pistons, &c.
    With the number of frickin' FIRES that we are constantly BURNING don't you think that things might warm up a little bit?

    Screw the talk about greenhouse gasses, &c. WHAT ABOUT ALL OF THE GAWDS DAMNED FIRES THAT WE'RE BURNING??????

  43. I'll wait for official word on that... by NimbleSquirrel · · Score: 1
    In the June 2008 Top 500 list, the Cray XT Jaguar was number 5 with 205 teraflop/s. By comparison, the number 1 was an IBM Roadrunner Bladecentre, with a mix of 6,562 Dual Core Opterons and 12,240 PowerXCell8i Cell Processors, housed in 278 cabinets. That got up to 1.026 petaflop/s.

    In June the Jaguar had 30,000 Quad Core Opterons, and now it has 45,000. The previous machine was an XT4, but the most recent update shows that 200 XT5 cabinets have been added to it. I have been unable to find how many cabinets the Jaguar has in total, but it seems that in June it had 313 (30,000 Opterons and 96 Opterons per cabinet). To me, the Jaguar seems to be two machines: the Cray XT4, and the Cray XT5. I'm also wary how increasing the number of processors by 50% yeilds an 800% performance increase. I'm going to wait until the official figures have been released on the 18th, when the next Top500 list comes out.

    If the Jaguar has had a performance increase, then I'd say the IBM machine would have had one too. It seems Cray are just fighting a war of attrition, trying to win back the supercomputing crown they held for so long (in the company's previous incarnations). They seem to be throwing processors at the problem. Yes there is more to supercomputers than processors (interconnects, switching, and memory management design are also vital ), but a 45,000 processor beast taking up 500+ cabinets is not a very elegant solution compared to a machine with 18,800 processors taking up only 278 cabinets (and arguably using far less power).

  44. Beep Beep by PingPongBoy · · Score: 2, Interesting

    In the not too distant future, we shall see a new Top 500 list. It just seems like yesterday that RoadRunner cracked the Petaflops barrier, and the whole world seems to have fallen on its ass in the interim. Banking failures, government bailouts, people losing their retirement portfolios. The irony is too much. Even as the computers get better, the answers that people need don't come fast enough.

    Then the light turned on for me. People in general, the people you see on the street going on their busy way to whatever, are mostly relying on "someone else" to come up with the answers. Most people have little confidence in their own ability to answer hard questions.

    Well, maybe things will turn around because of the power of supercomputers. It would be about time, wouldn't it? Here's how it may play out. Supercomputers so far, good as they are, serve up expensive results, so they are applied to difficult problems that are useful but far removed from everyday life.

    As supercomputer clock cycles become more abundant, researchers can apply them to do more mundane things that the unwashed can relate to. The result could be revolutionary. People who have always aspired to some inconsequential achievement that requires some expertise or training may suddenly have access to highly instructive supercomputer-generated procedures that explain both how and why. Not only will people become more expert do-it-yourselfers, but robots will become far versatile, with amazing repertoires.

    Crossing the petaflops barrier may be sufficient psychological incentive for people to request that governments begin to make supercomputing infrastructure available for public consumption, like roads and other services. Certainly, exciting times are comiing.

    --
    Know your pads. One time pad: good for cryptography. Two timing pad: where to take your mistress.
    1. Re:Beep Beep by Anonymous Coward · · Score: 0

      RoadRunner? Heavens, it seems like yesterday that everyone was in awe about the Earth Simulator, which actually remained in the #1 spot for a couple of lists - these days, it's on #49 in the current list (and it'll drop further when the next one is announced in three days, of course). Little more than an also-ran, really - it's about in the same performance range as the fastest supercomputer in the Netherlands. (Nothing against the Netherlands, BTW. Lovely little country.)

  45. Awesome, by nog_lorp · · Score: 1

    Will it run Cybermorph?

    1. Re:Awesome, by Polarina · · Score: 1

      What won't it run?

  46. How about... by evariste.galois · · Score: 1

    creating better algorithms? Or at least educating a little bit all non-CS scientists about performance and optimization? If you go over the climate-prediction loop many many times, you should consider some caching..

    1. Re:How about... by Ambitwistor · · Score: 2, Informative

      creating better algorithms? Or at least educating a little bit all non-CS scientists about performance and optimization?

      The guys who work on the the dynamic cores for the biggest climate models (NCAR, GFDL, NASA, etc.) do world class numerical hydrodynamics. Maybe not quite on par with the nuke guys at, say, Sandia, but pretty good. And they do hire programmers and numerical methods people to do algorithm design, optimization, and parallelization. They're cutting edge in terms of grid solver algorithms for these sorts of problems. There are lots of complications from irregular topography, coupling between atmosphere, ocean, biosphere, and cryosphere, etc.

      If you go over the climate-prediction loop many many times, you should consider some caching..

      Caching does little good, because none of the grid cells have the same value after each time step. Some things just require a supercomputer. There's a reason why people use supercomputers for big 3D fluid dynamics simulations (nuclear explosions, virtual wind tunnels for aerospace, climate/weather models, etc.)

  47. no one mentioned? by Anonymous Coward · · Score: 0

    It runs Linux, whee!

  48. It runs Linux... by mok000 · · Score: 1

    Heh. Worlds fastest computer runs Linux.

  49. Skynet is still the best by Anonymous Coward · · Score: 0

    Cause they have Arnold!

  50. Possibilities by Everlife · · Score: 1

    Yeah, but can it run Duke Nukem?

    1. Re:Possibilities by strjms72 · · Score: 1

      that would be something :P

  51. Supercomputer by tonnywilliams · · Score: 1

    Just five months after IBM's hybrid Roadrunner became the first supercomputer to break the lofty petaflop barrier, a second, more traditional machine has made the same leap .And at least one industry watcher said the move of Cray's XT Jaguar supercomputer -- with the help of a $100 million upgrade -- into the petaflop realm is swinging the doors wide open for other systems that are on the verge of following it to a new level of power and speed. --------------- Tonnywilliams Your DUI News

  52. Beowolf anyone.. by AnonNoMore · · Score: 1

    I'm amazed.. I mean this is Slashdot, How could an article about some kind of computer get posted without the obligatory: Imagine a Beowolf cluster of these things.. :)

  53. Hello World! by godztempus · · Score: 1

    That would be my program if I had a chance to run something on it. I would make sure it ran on every core too.