Slashdot Mirror


Japan's New Supercomputing Toy

deman1985 writes "As reported by UPI, Japan has unveiled their fastest supercomputer yet. Assembled from Hitachi and IBM components, the new system sports total performance around 59 trillion calculations per second and comes at a cool 5-year lease price of $30 million. Pictures of the beast can be found at Mainichi Daily News."

190 comments

  1. FP by Enzero · · Score: 0, Offtopic

    FP

    1. Re:FP by Anonymous Coward · · Score: 0

      59 trillion of them! Per second!

  2. I wonder by bherman · · Score: 1

    Is it getting the most of that computing power by running Windows?

    --
    Error: Sig not found.
    1. Re:I wonder by Anonymous Coward · · Score: 0

      Just in case anyone does not know, Blue Gene and Blue Gene/L both run Linux.

    2. Re:I wonder by Cygfrydd · · Score: 2, Funny

      Reportedly, it is just fast enough to use Windows Vista’s Aero Glass GUI.

    3. Re:I wonder by Anonymous Coward · · Score: 0

      /golfclap

    4. Re:I wonder by Anonymous Coward · · Score: 0

      funny

    5. Re:I wonder by Anonymous Coward · · Score: 0

      not

    6. Re:I wonder by Anonymous Coward · · Score: 0

      i would think that powerPC/ intel/ amd based clusters would be slower than some sort of streaming processing core like a GPU, with the progress that has been made in General Purpose Programming on the GPU i would think that you could make a super scalar graphics card cluster that is networked together with either copper gigabit, or optical 10 gigabit connections.

      from my calc, it would use far less power, cost less to build and would be more parallel than a risc/cisc based design.

      the advantage would trend towards bio sims, weather sims, physics sims, where finite particles of information need to interoperate on a massive scale.

      jDavid.net "i have enough passwords that are not written down to remember"

    7. Re:I wonder by PitaBred · · Score: 1

      Except for the fact that precision on GPU's is highly limited. That's why most of these grand ideas are non-starters for scientific applications. GPU's have their place, but high precision mathematics isn't it right now.

    8. Re:I wonder by TinyManCan · · Score: 1

      You are technically incorrect on this. The Blue Gene I/O Nodes use Linux. The compute nodes run a custom OS. So they use Linux for accessing hardware devices, and their own custom OS for actually doing the work.

  3. Yes.... by JoeLinux · · Score: 0, Redundant

    But imagine a a beowulf cluster of these.

    *sigh* I miss when that was popular...I was in college, dating a total bitch, living off of ramen, playing CS until my grades started to suffer, and getting four hours of sleep a night...good times, good times.

    1. Re:Yes.... by j79 · · Score: 2, Insightful

      *sigh* I miss when that was popular...I was in college, dating a total bitch, living off of ramen, playing CS until my grades started to suffer, and getting four hours of sleep a night...good times, good times.

      Shit. That's my life right now, and trust me...it ain't good times...

      Then again, maybe a few years down the road, when I have a shit job, married to a total bitch, living off of ramen, and still play games till the wee hours in the morning, I'll be able to reflect and think, "yeah...good times...good times..."

      Wow. That just made me even more depressed.

      *SIGH*

    2. Re:Yes.... by Anonymous Coward · · Score: 0

      Maybe if you stopped referring to your SO as "bitch", you wouldn't be so unhappy?

      Just a thought...

    3. Re:Yes.... by IdleTime · · Score: 1

      Why do you tell people to lie?

      --
      If you mod me down, I *will* introduce you to my sister!
    4. Re:Yes.... by danath333 · · Score: 0

      i hear you man, good times indeed. amen.

  4. Beowulf Cluster by Nazmun · · Score: 0, Redundant

    Imagine a beowulf cluster of these. And also does this run linux?!?!?

    There, i've said it...you know someone would have!

    --
    Hmmm... Pie...
    1. Re:Beowulf Cluster by Anonymous+Rockstar · · Score: 0, Offtopic

      and don't forget the ever popular... Now I can run Longhorn/Vista.

      --

  5. I for one... by JDSalinger · · Score: 0, Redundant

    I for one welcome our Japanese Supercomputing Toy overlord! I also welcome Godzilla to fight this Supercomputer in an epic battle (Pictures of the beast can be found at a random website http://www.xofacto.com/justin20/shinzen-godzilla-6 2.jpg). I also welcome females back to my crib.

  6. Fastest? by leipzig3 · · Score: 1

    According to top500.org, the fastest computer is an IBM Blue Gene/L with 280 TeraFlops. This Japanese team would have been #1 about a year ago.

    1. Re:Fastest? by WeAzElMaN · · Score: 1

      The article didn't say that it was the fastest in the world, but merely the fastest to yet be built by the Japanese.

      Yes, Blue Gene/L still reigns supreme.

      -WeAz

    2. Re:Fastest? by Digi-John · · Score: 1

      According to top500.org, the fastest computer is an IBM Blue Gene/L with 280 TeraFlops. This Japanese team would have been #1 about a year ago.

      What?!?!111one It's from JAPAN, it must be the fastest! Don't tell me your evil lies, Japan is the land of super technology, games, and girls with cat ears! NIHON BANZAI!

      --
      Klingon programs don't timeshare, they battle for supremacy.
  7. And Yet by Herkum01 · · Score: 0, Troll

    It still cannot run Windows Vista...

    1. Re:And Yet by PishiGorbeh · · Score: 0, Offtopic

      I might run Vista but is it fast enough to run Vista with Areo?

    2. Re:And Yet by Anonymous Coward · · Score: 0

      Nor serve more than 10 ruby on rails pages per second.

    3. Re:And Yet by dvhh · · Score: 1

      And OSX hasn't been hacked yet to run on these .....

  8. ask public? by tomstdenis · · Score: 2, Interesting

    They're gonna ask the public for research themes? ... AFTER THEY BOUGHT IT???

    I'd love to see this from top500.org

    name,where,how many processors,average FLOPS,max FLOPS,***actually being used FLOPS***

    Then sort it based on the latter. :-)

    Tom

    --
    Someday, I'll have a real sig.
    1. Re:ask public? by Anonymous Coward · · Score: 0

      Average FLOPS? You don't know what you're talking about.

    2. Re:ask public? by the_humeister · · Score: 1

      Yes, hopefully the response won't be something like "Prove that creationism is true." Boy that be a fun one to calculate...

      Also, what's the point in having the Hitachi around when the IBM sitting right next to it is 25 times faster?

    3. Re:ask public? by Tower · · Score: 1

      Well, from the article, the main part of this is a Blue Gene system (57.3 Teraflops according to the article).... so looking at http://top500.org/lists/2005/11/TOP10_Nov2005.pdf that would put it in 4th place... if you add the Hitachi flogs in there.... 4th place.

      --
      "It's tough to be bilingual when you get hit in the head."
    4. Re:ask public? by luckyguesser · · Score: 1

      how about 'What is the answer to life, the universe, and everything?'.... i would think that would be the obvious question :P

      --


      The power of Christ compiles you.
      A Random Blog
  9. Uses? by Douglas+Simmons · · Score: 1
    That's a lot of calculations per second, but other than satisfying man's desire to have the biggest or fastest thing on the block, what possible uses of real value, other than decryption, could this thing bring to the table? Quicker searching for prime numbers? Weather modeling? SETI@Home nonsense?

    And does anyone have an update on the Jap's supersonic jet project? Last story I remember was a model crashing in Australia. Go Japan!

    1. Re:Uses? by Rob+T+Firefly · · Score: 1

      You're Bill Gates and I claim my five pounds.

    2. Re:Uses? by muhgcee · · Score: 1

      What uses? TFA said particle accelerator research. So they're studying physics. What's wrong with that?

    3. Re:Uses? by Douglas+Simmons · · Score: 1

      Fuck TFA.

    4. Re:Uses? by everphilski · · Score: 1

      EQ2 at 30FPS :P

      (i kid, i kid)

    5. Re:Uses? by Mercaptan · · Score: 1

      Biological simulations.

      Protein folding, modeling cell machinery, and simulations of other biological systems at the molecular level. Think about the number of calculations involved in modeling the interactions between a few million atoms in something as simple as the ribosome. Now imagine adding the water and solute environment that surrounds these sorts of molecules. Oh, you could ignore the water and do the simulation in a vacuum, but let's remember that a driving force in protein conformation is hydrophobicity. And if you want to try this over time, say even a few nanoseconds, then start multiplying baby.

      So yeah, there are some good uses.

      --
      -- "Sucks to your ass-mar"
    6. Re:Uses? by cdrudge · · Score: 1

      I beleive traditional uses is in partical physics and nuclear testing, among other things. Knowing what happens when you slam two particals together, along with the reaction to the surrounding enviroment, takes an extrodinary amount of calculations.

    7. Re:Uses? by Anonymous Coward · · Score: 0

      You have to remember, these are Japanese men... they have to do SOMETHING to compensate for their tiny penises!

    8. Re:Uses? by canning · · Score: 1

      I guess if you need a supercomputer why not build the fastest?

      --
      I love the smell of Karma in the morning
    9. Re:Uses? by Jeff+DeMaagd · · Score: 1

      Super computers are used in weather simulations, weapon design simulations, energy grid and so on. Louisiana's vulnerabilities to hurricanes was learned around 1999 because of a supercomputer simulations, the problem is that politicians apparently chose to ignore the threat. One man is trying to use supercomputers to help determine if there was a way to affect a hurricane by seeing if it could be affected with weather modification tools when it starts to form.

    10. Re:Uses? by afidel · · Score: 1

      Protein folding. Proteonics makes the human genome project seem like an insignificant ant next to a Boing, both in terms of required CPU power and in regards to potential benifit to the human race.

      --
      There are 4 boxes to use in the defense of liberty: soap, ballot, jury, ammo. Use in that order. Starting now.
    11. Re:Uses? by flimflammer · · Score: 1

      I'm curious to how you could really ask that question. There are numerous reasons to have such systems with such speeds. Many of which are previously mentioned.

    12. Re:Uses? by Attila+the+Bun · · Score: 1
      other than satisfying man's desire to have the biggest or fastest thing on the block, what possible uses of real value, other than decryption, could this thing bring to the table?

      If only there were an easy way to find out. If only the story had a link to a web page with more information. Sigh...

    13. Re:Uses? by Anonymous Coward · · Score: 0

      Yeah, destroying the American car industry was a fun hobby, but it's clearly time to move on to greener pastures.

  10. Give it 5 years by airlynx · · Score: 1

    Give it five years and it'll be a commercially available laptop, ten at the most.

    --
    I got into Linux for the free beer, but nobody seems to have any
    1. Re:Give it 5 years by woah · · Score: 1

      Good luck with that. Moore's law is dead.

    2. Re:Give it 5 years by TinyManCan · · Score: 1
      The number of transistors is still doubling at an hectic pace. Maybe you are confused (as are most people) about what Moore's law actually is.

      Now with mulit-core processors becoming the norm, and 32 cores per CPU not that far off, it is easy to see that Moore's law is going to keep going strong for at least 5 more years, if not more.

      Since these machines are already distributed multi-node systems, a single cpu with 32 cores is going to be generally faster than 32 CPUs in a distributed node configuration. So basically, we may not have it in our latops, but workstations and low-end servers maybe. Higher end servers certainly.

  11. I suggest size cap on these stats... by syslog · · Score: 1

    You know, I think there should be a kind of "size cap" on these stats. A computer should be ranked higher if it can squeeze more performance out of so many cm2 worth of die or something. Otherwise you can just keep making computers more powerful by just adding more and more nodes. That does not excite me *shrug*. -naeem

    1. Re:I suggest size cap on these stats... by muhgcee · · Score: 1

      I think the more appropriate statistic would be computations per watt. Because land can be very cheap, but electricity doesn't have nearly the same variance across different parts of the world/country.

    2. Re:I suggest size cap on these stats... by syslog · · Score: 1
      Agreed. The only thing I wanted to stress is that there has to be *some* reasonable sandbox within which the compute power of these machines is measured. Otherwise just keep tacking those nodes on to move higher and higher on this most useless of rankings... :)

      -naeem

    3. Re:I suggest size cap on these stats... by tetabiate · · Score: 1

      I totally agree, There are in fact some physical and mathematical problems involving correlations which might not be easy to parallelize or cannot be parallelized at all.

    4. Re:I suggest size cap on these stats... by kukickface · · Score: 1

      Doesn't bounding an abstract entity such as computation to a physical one such as volume or power consumption seem irrelevant?

  12. Fastest Supercomputer Yet by digitaldc · · Score: 1

    Now what to do with it?

    How about installing one over at Slashdot HQ?
    You guys need it for all the people who keep missing their chance at getting the first post.

    --
    He who knows best knows how little he knows. - Thomas Jefferson
  13. teraflops by wesw02 · · Score: 1

    Does 59 trillion calculations approximately equal 59 teraflops?

    1. Re:teraflops by Spy+der+Mann · · Score: 1

      Does 59 trillion calculations approximately equal 59 teraflops?

      Good point. I live in Mexico, and here "illions" are measured in 10^6 units. So here, a billion is 10^12, a trillion is 10^18, etc.

      And actually I don't know how it's handled in different countries, so yes, it's confusing. Using mega,giga,tera is much more specific and doesn't lead to confusions.

    2. Re:teraflops by oliverthered · · Score: 1

      Same here in Europe, all US mesurements seem to be a bit screwed (hence the US gallon etc...) there is a milliard which is what the US call a billion.

      --
      thank God the internet isn't a human right.
    3. Re:teraflops by Secrity · · Score: 2, Informative

      59 trillion calculations a second equals exactly 59 teraflops, of which, 57.3 of the 59 teraflops is from a smallish IBM Blue Gene.

      Lawrence Livermore National Laboratory has an IBM Blue Gene that does 280.6 teraflops or 280.6 trillion calculations a second.

    4. Re:teraflops by ciroknight · · Score: 1

      Not if they're integer operations. Then its 59 teraiops.

      I still prefer 59 teracyclops though.

      --
      "Victory means exit strategy, and it's important for the President to explain to us what the exit strategy is." G.W.Bush
    5. Re:teraflops by Main+Gauche · · Score: 1

      "Same here in Europe, all US mesurements seem to be a bit screwed (hence the US gallon etc...) there is a milliard which is what the US call a billion."

      Yes, but to avoid confusion, we both have billiards.

    6. Re:teraflops by Anonymous Coward · · Score: 0

      "280.6 teraflops or 280.6 trillion calculations a second."

      *nerdgasms*

    7. Re:teraflops by jbNet · · Score: 0

      wouldn't a milliard be 1/1000th of a yard?

    8. Re:teraflops by oliverthered · · Score: 1

      It's the name given to a tough bunch of millipedes.

      --
      thank God the internet isn't a human right.
  14. nr 6? by B.Stolk · · Score: 1

    Isn't that nr 6 in the world?
    The list is here.

    --
    http://www.stolk.org/tlctc
  15. IBM Blue Gene Project by nharmon · · Score: 1
    1. Re:IBM Blue Gene Project by Locke2005 · · Score: 1

      Yes, it runs Linux! Well, sort of... the I/O processors run Linux; the actual compute nodes run a proprietary low-overhead executive. Linux is a general purpose OS, and interrupting CPUs for system processes and timer ticks has a huge cumulative impact on highly parallel tasks. Ideally, one would want the compute nodes to get a chunk of work and work on nothing else until that chunk of work is finished, i.e. preemptive multitasking is actually a liability for massively parallel machines.

      --
      I've abandoned my search for truth; now I'm just looking for some useful delusions.
  16. Japan's fastest? by Anonymous Coward · · Score: 0

    The supercomputer, consisting of two systems -- Hitachi's multipurpose supercomputer with a peak performance of 2.15 terra flops and IBM Japan's Blue Gene Solution with a peak performance of 57.3 terra flops -- is capable of making about 59 trillion calculations per second, the Mainichi Shimbun reported Wednesday.

    Err, I know it comes from IBM Japan, but since the bulk of the horsepower is coming from an IBM Blue Gene, does this really qualify as meaningful with relation to Japan? I think they'd be much more interested in Hitachi coming out with their own unit that could topple the IBM's.

  17. slanted? by ejamie · · Score: 1

    Interesting slanted design of the racks.

    Is there a design reasons for that (air flow, etc)?

    Or is this marketing wanting to be "different"?

    --
    Hey! Stop copying my sig!!! Stop copying my sig!!! Stop copying my sig!!! Stop copying my sig!!!
    1. Re:slanted? by Anonymous Coward · · Score: 0

      It's slanted because it was installed in Asia, it would have been round for the US install

    2. Re:slanted? by jo42 · · Score: 1

      Yes, it would go faster if they angled it even more. And added a "Type R" sticker with soup can size pipes.

    3. Re:slanted? by Anonymous Coward · · Score: 0

      It needs a spoiler, and should probably be lower to the ground.

  18. This thing's not so fast... by prisoner-of-enigma · · Score: 1

    ...but imagine a Beowulf cluster of them!

    Hey, somebody had to say it!

    --
    In the end they will lay their freedom at our feet and say to us, Make us your slaves, but feed us. - Fyodor Dostoyevsky
    1. Re:This thing's not so fast... by cdrudge · · Score: 1

      You know, these comments were funny. In 1998. 8 years later, their dumb. No. Seriously. No one is laughing.

    2. Re:This thing's not so fast... by prisoner-of-enigma · · Score: 1

      8 years later, their dumb.

      And eight years later you still haven't discovered the difference between "their" and "they're." You're a poster child for grammatical stupidity all by yourself, so remind me again why you're casting stones?

      --
      In the end they will lay their freedom at our feet and say to us, Make us your slaves, but feed us. - Fyodor Dostoyevsky
    3. Re:This thing's not so fast... by Jokerz17 · · Score: 1

      You know, these comments were funny. In 1998. 8 years later, their dumb. No. Seriously. No one is laughing.

      I thought it was funny.

      Seriously... imagine a beowolf cluster of these.

    4. Re:This thing's not so fast... by cdrudge · · Score: 1

      Yeah, yeah, yeah. There is a difference thought between being a Humorless Reposted Joke Nazi and a Grammar Nazi. I used the wrong form of their/there/they're. I stand corrected. Please don't stone me.

      My original point still remains.

  19. Proposal by phoric · · Score: 1

    "The institute will ask the public to propose specific themes of research activities using the supercomputer system." Seti@Home! Duh...

  20. Toy by muhgcee · · Score: 1

    "Toy" seems pretty appropriate here. They seem to have bought it without fully knowing what they were going to do with it yet: "The institute will ask the public to propose specific themes of research activities using the supercomputer system."

    1. Re:Toy by 9Nails · · Score: 1

      Call me silly, but at least I'm not the only one who was thinking that thought!

      We're going to spend several tens of millions of dollars on this new Super Computer! Wooot! Ummm, what do need it for?

      Here's an idea; Hey Japan, how about an Excel Spreadsheet that calculates how much money you lose per second with that acquisition?

  21. Its real use by BecomingLumberg · · Score: 1

    is to allow for more FF-XI servers. Duh.

    --
    If a nation expects to be ignorant and free, in a state of civilization, it expects what never was and never will be.-TJ
    1. Re:Its real use by Anonymous Coward · · Score: 0

      Funny you should mention that. EVE Online just installed a new cluster consisting of 70 dual Opteron nodes. Sure, it's no Earth weather simulator, but it's certainly in the running for having a little sticker labelled, 'Supercomputer!' stuck on it.

      And as subscribers continue to rise, they'll undoubtedly reach the point where if they wanted to, they could easily can the game and run, say, nuclear weapons simulations. :p

      It's interesting to see how far client-server games have come. Yesterday, it was some guy living down the block with a 486. Tomorrow, it'll be top of the line, world-class supercomputers.

  22. Using PowerPC processors too by kuwan · · Score: 1

    Processor - PowerPC 440 700MHz; two per compute node - Lowpower allows dense packaging; better processor-memory balance

    Not particularly powerful CPUs individually, but I guess if you cram enough of them together it adds up.

    1. Re:Using PowerPC processors too by Jeff+DeMaagd · · Score: 1

      The last part you quoted was very important. Supercomputers often spend more time waiting for memory than they do with actual computing. The CPU doesn't need to be so fast for this.

    2. Re:Using PowerPC processors too by Kaldaien · · Score: 2, Interesting

      Indeed, that is one of the reasons IBM is so proud of their new Cell architecture. It was designed to reduce the latency between CPU and RAM, perhaps more out of necessity than by choice. IBM wanted to reduce the complexity of the processor by moving a lot of the out-of-order-execution, register renaming, branch prediction, etc... logic off the silicon and into the compiler. Transmetta tried the same thing in the past, but found it only compounded memory latency issues. It works out for the Cell architecture, however, since it is designed around parallelism -- with a lot of the complexity dumped on the compiler, there is a lot more room on the die for more SPFs, cache and other logic. It should be interesting to see if the theory actually works beyond paper... it sounds like playing musical chairs with bottlenecks and praying compiler developers will do the really hard work. ;)

    3. Re:Using PowerPC processors too by Kaldaien · · Score: 1

      SPEs, not SPFs :)

    4. Re:Using PowerPC processors too by flaming-opus · · Score: 1

      Which is all well and good, except that the cell architecture gives the spe's really fast access to a very small pool of memory, and then a decent, but not astounding access to the rest of memory through the master core. This works very well for doing a lot of number crunching on a small bit of digital media data. However, there are a lot of supercomputing problems that are bounded by the amount of memory addressable by each processor. Even Blue Gene's quarter gig per node is a problem for many codes.

      If the successor to blue gene runs with some cell-like processor, which I suspect it will, they'll have to do more than just drop PS3 cells onto a board and run with it. There will have to be some reballancing of the internal/external memory bandwidth.

  23. Here's one... by from+mars · · Score: 0

    Well, the answer to life, the universe and everything still hasn't been answered....

  24. What's happened to the moderation system??? by advocate_one · · Score: 3, Interesting

    Have we had another moderation system failure... the number of posts making it to my +3 browsing has dropped dramatically in the last two days... I'm expecting there to be almost minimal moderation after today and it to be a general trollfest again by Friday...

    --
    Donald 'Duck' Dunn: We had a band powerful enough to turn goat piss into gasoline.
    1. Re:What's happened to the moderation system??? by ZachPruckowski · · Score: 1

      I've noticed the same thing. there seem to be a lot of down-mods burning the mod-points. I've seen stories out for hours with all of 0 Score: 4s or 5s

    2. Re:What's happened to the moderation system??? by toxicity69 · · Score: 1

      I noticed that also, and attributed it to most of /. using up their karma modding everyone in the Monty Python poll a +5 interesting/insightful/funny.

    3. Re:What's happened to the moderation system??? by demachina · · Score: 2, Insightful

      Having read the replies so far to this thread there simply isn't anything worthy of being modded up, in particular there is a steady stream of predictable Beowulf comments, which just aren't funny or worth posting anymore.

      At least for this thread, I'd say its not the moderation system failing, there just isn't anyone with anything intelligent to say posting anything.

      Slashdot posters seem to be the thing cratering.

      Articles on IBM throwing together another giant collection of CPU's for someone with money to burn really aren't very interesting anymore either. If it was an absolute record that would be something but this system is already behind similar IBM systems already in U.S. labs. If it was a revolutionary new architecture that would be something but this isn't.

      Governments somehow seem to think they can garner prestige by assembling these white elephants that tend to be obsolete before a 5 year lease is up. Me personally I think they should be focusing more on developing new and interesting software that does useful things on somewhat smaller and more practical systems. Well done and useful software has a lot longer lifespan than thousands of CPU's that will be obsolete in a couple years.

      I'm sure climate modelling, molecular modelling, CFD etc. might find a good home on this machine but the machines themselves just really aren't that newsworthy any more except to the IBM marketing department and the governments engaged in the penis size contest.

      --
      @de_machina
    4. Re:What's happened to the moderation system??? by TheLetterPsy · · Score: 1

      I've noticed the same. At first I thought it was because I changed my comments preferences but when I set them back to the way they were, no change.

      It used to be that browsing at +3, you would get about 10-20% of the total number of comments (e.g. 70-120 comments +3 or higher when total comments = 700). I thought at first (in typical slashdotter fashion) that there was a massive conspiracy of rogue mods who just downmodded at will. But looking at the mod scores, it seems like it's just that there aren't as many mods total or people just don't care for moderating any more (all of a sudden).

      Who knows? Maybe we'll have one of those CmdrTaco navel/naval gazing posts again . . . one can only hope.

    5. Re:What's happened to the moderation system??? by shadowbearer · · Score: 1


        Is there anything a supercomputer can do that thousands of clustered machines can't? (Honest question)

      SB

      --
      It's old. The more humans I meet, the more I like my cats. At least they are honest.
    6. Re:What's happened to the moderation system??? by pimpimpim · · Score: 1
      I am not a supercomputerbuilder, but the IBM "supercomputers" are actually just "clustered machines" as far as I know. But that still doesn't make them a "cluster".

      It's probably a question of definition. The CTO of Cray answered this question some time ago. Now he might be biased towards supercomputers, but the way I read it there is not much difference at the moment except that supercomputers are specifically built to have fast and broad interconnects between CPUs and an optimal way to share CPU. A cluster is built out of standard parts, therefore has less shared memory, slower inter-CPU connections etc.

      Of course the supercomputer will be better, but also a factor 10 more expensive when comparing same amount of single CPU power. But I would say it's worth it, the simulations don't gain much from parallel runs on a simple ethernet-cluster, but when you go to stuff like myrinet connects it really speeds like hell :)

      --
      molmod.com - computing tips from a molecular modeling
  25. how long before a singularity? by Sub+Zero+992 · · Score: 1

    Hi,

    Given Moore's Law, and given increasing performance gains in computer architecture and new work on algorithms, how likely is it that one of these days one of these machines (or one of their exponentially more powerful progeny) bootstrap themselves into a "Singularity", an AI which at the point of self-awareness becomes almost instantaneously god-like?

    I know that this has been the stuff of Science - Fiction wet dreams for decades, but will this old idea - like so many other ideas first found in science-fiction - one day become reality? And would the God-Head be dictatorial in its nihilistic contempt of its creators, or perhaps more Frankenstinian in its attempts to be accepted by humanity?

    --
    They who would give up an essential liberty for temporary security, deserve neither liberty or security - Ben Franklin
    1. Re:how long before a singularity? by sandmaninator · · Score: 1

      Given Moore's Law, and given increasing performance gains in computer architecture and new work on algorithms, how likely is it that one of these days one of these machines (or one of their exponentially more powerful progeny) bootstrap themselves into a "Singularity", an AI which at the point of self-awareness becomes almost instantaneously god-like?

      No.

    2. Re:how long before a singularity? by Anonymous Coward · · Score: 0

      Simple answer. Never. Computers in their current form are only going to run the software we program, and the speed of a computer doesn't change the software -- just how fast it is executed. At present we don't have the faintest clue how to make an 'intelligence software' that is intelligent in the way a human is (*cough*) intelligent. We aren't even sure if it is possible to achieve intelligence using this approach. And sorry, spilling your softdrink into your computer isn't going to make it intellgent either except in the movies ... even if you did it to a billion computers a second until the end of time. Intelligent computers are going to require great advances in our understanding of intelligence, perhaps a completely different kind of computer, and certainly the acculmulated hard work on the part of many, many people.

    3. Re:how long before a singularity? by Dark_MadMax666 · · Score: 1

      It doesnt "magically" become self -aware. Strap 100 times more procesisng power doing yet another weather-partcile-whatever computational intesive taske and it still not be self-aware. It has to be specificaly designed and desgnated for this task. I havent heard any supercompturs used for doing anything AI specific yet . I am sure they will some day , but AI wont just emerge from paricle analyzer .

    4. Re:how long before a singularity? by joss · · Score: 1

      Computers are all the same - some of them have more memory than others and some of them are speedier than others but they are functionally the same. Given enough time and memory, any computer can simulate any other computer. Intelligence and speed are separate concepts. Imagine we had a chat with some aliens who lived 10 light years away. There answers would be no less intelligent even though we had to wait 20 years to receive a reply.

      So, intelligence is no more likely to emerge all of a sudden from the latest supercomputer than from that rusty old pentium you've been using as a firewall. Either human brains are not Turing machines, or the hardware is sufficient to host super intelligence already. Either way, the software has got a long way to go. Kurzweil believes we'll be there in 30 years, but his arguments about software advances [and brain reverse engineering] were somewhat unconvincing.

      --
      http://rareformnewmedia.com/
    5. Re:how long before a singularity? by Durf · · Score: 1

      It would never be accepted by humanity. The first words out of its mouth would be something like "Russian brides v|ag ra! pok3r rooms onl in3 ca$in0!" and humans hate that kind of gibberish.

  26. Must be good at math... by Ixne · · Score: 1
    The supercomputer, consisting of two systems -- Hitachi's multipurpose supercomputer with a peak performance of 2.15 terra flops and IBM Japan's Blue Gene Solution with a peak performance of 57.3 terra flops -- is capable of making about 59 trillion calculations per second...

    You don't say. I wonder if they put those numbers into the machine(s) to get that sum...
  27. Off-Topic by wanerious · · Score: 1

    Ok, I haven't been keeping up --- this Japanese machine is, I gather, massively parallel. Suppose I wanted to find out which single processor was the speed king for floating point calculations. Is it as simple as sorting for the highest number on SpecFP2000?

    1. Re:Off-Topic by Anonymous Coward · · Score: 0

      700 mhz powerpc440 processor. 2 per node.

    2. Re:Off-Topic by flaming-opus · · Score: 1

      First you would have to define what you mean by "single processor". The fastest addressable processor is probably fujitsu's VPP series node, but it's not really one processor, it's 8 IBM power4+ processors ganged together with a vector coupling facility. The fastest hpc processor that sits on a single chip is the cray X1e msp chip, which has 2 18GFlop cores per chip. This of course ignores specialized chips like DSPs, most notably the IBM cell chip. Cell has a much higher flops performance than any chips used in HPC systems, but it's difficult to call it a single processor, as it runs from 9 instruction streams at once, and only does 32bit math anyway.

      If you're asking what the fastest 64bit, superscallar microprocessor is, for general floating point codes, it's probably itanium, but maybe power5, depending on what you're doing, and how much optimization you've done.

    3. Re:Off-Topic by flaming-opus · · Score: 1

      I'm sorry. I got that wrong. It's not the fujitsu vpp (which is no longer being made), but rather the hitachi sr11000, which gangs together a bunch of power4's into a vector processor.

  28. 5-year-lease by RealProgrammer · · Score: 1

    Under Moore's Law, the price of computing power halves every 18 months. So that means in 5 years, the price ought to be about 1/(2^(5/1.5)) =~ a tenth of that.

    That means in five years I'll be able to afford it on my desktop for about what I make an hour.

    Woohoo!

    I hope it comes with a better mouse. I have one of those mechanical ones, and it keeps getting granola in it.

    --
    sigs, as if you care.
    1. Re:5-year-lease by PastAustin · · Score: 0
      I hope it comes with a better mouse. I have one of those mechanical ones, and it keeps getting granola in it.


      Posts like this are why I still read slashdot.


      Thank you!
      --
      Firefox 2.0 - Spell Rightly.
    2. Re:5-year-lease by thogard · · Score: 1

      No, mores law claims the transistor density (or now the data density) doubles ever 18 mo. It turns out that increases the problems you can solve in a non linear way. A great example of this is the hardware multiplier. Once transistor density got tot he point where you could put a hardware multiplier on the chip, you could do a 16 bit multiply about 60 times faster than using the add/shift technique so you do do far more complex work. There are also massive gains in DNA sequencing once you could build a machine that could hold an entire DNA strand in memory so it didn't have to be doing comparisons off disk which resulted in a several million time speedup.

  29. This contains a BlueGene implementation by tubbtubb · · Score: 1

    True, the biggest BlueGene/L implementation does best this number. Also interesting to note, this thing has BlueGene in it: " The supercomputer, consisting of two systems -- Hitachi's multipurpose supercomputer with a peak performance of 2.15 terra flops and IBM Japan's Blue Gene Solution with a peak performance of 57.3 terra flops -- is capable of making about 59 trillion calculations per second, the Mainichi Shimbun reported Wednesday. "

    1. Re:This contains a BlueGene implementation by Pollardito · · Score: 1
      Hitachi's multipurpose supercomputer with a peak performance of 2.15 terra flops and IBM Japan's Blue Gene Solution with a peak performance of 57.3 terra flops -- is capable of making about 59 trillion calculations per second, the Mainichi Shimbun reported Wednesday.
      i wonder how many nanoseconds it took this new supercomputer to add 2.15 to 57.3 and round that off to 59
    2. Re:This contains a BlueGene implementation by Anonymous Coward · · Score: 0

      I hope they didn't blow a vacuum tube :0

  30. Like the name... KEK by BitwizeGHC · · Score: 3, Funny

    In other news, South Korea unveiled its new supercomputer: KEKEKE ^____^

    --
    N4st0r, trixx0r h0bb1tz0rz! Th3y st0l3 0ur pr3c10uzz!
  31. Geekdom by bookemdano63 · · Score: 1

    How sad is it that I was thrilled they included a link to a picture of it?

  32. Can We Please Institue an Intelligence Filter? by raftpeople · · Score: 1

    And how did the parent get a score of 2?

  33. NOT a Toy! by ObsessiveMathsFreak · · Score: 1

    It's not a toy OK! It just's able to run Counterstrike at really high framerates because scienctific simulations and game mechanics are very similar operations...

    Now if you'll excuse me, my aimbots need seeing to.

    --
    May the Maths Be with you!
  34. Slashdotted by G3ckoG33k · · Score: 1

    I just wait for that beast to get slahsdotted, any minute now. Wait, they didn't run the web server on that one, did they? High energy accelerator research... Hrrmmmppff..

  35. We're almost there... by suitepotato · · Score: 1

    ...we almost have a computer that can start Adobe applications in less than one full work day.

    --
    If my grammar and spelling are off, I am [distracted/tired/careless] (take your pick)
  36. yup, cooling by tubbtubb · · Score: 1

    Yeah, its for air flow.
    See the large ppt presentation here:
    http://www.research.ibm.com/bluegene/briefing_day. html
    Power, Packaging and Cooling, slide 20, I think.
    Basically, the ducts are larger where the airflow is greater.

  37. PS3 by dasare1503 · · Score: 0, Offtopic

    What they don't tell you is that the PlayStation 3 is expected to have the power of TWO of these babies. Also, it will model the weather in your neighborhood down to the atomic molecule scale, WHILE putting out fifteenteen HD video streams. Sony PS3 department was heard to say "We are not competing against that new supercomputer. It is too slow."

  38. 2006 supercomputer = ten teraflops by peter303 · · Score: 1

    I think when the lists are released this year, the top one hundred computers will be in the 10 to 300 teraflop range. With the cell CPU chipset peaking at a quarter-teraflop, one teraflop is merely high performance these days.

  39. Apple will use this in their next Mac Mini by TheSkepticalOptimist · · Score: 1

    Their tag line will be "Finally, something fun to do with 59 trillion calculations per second!". But nobody will buy it still because there really isn't anything fun about Mac what with virtually no game support.

    --
    I haven't thought of anything clever to put here, but then again most of you haven't either.
  40. Why have the Hitachi? by Anonymous Coward · · Score: 0

    What use if the Hitachi supercomputer if it only adds a mere 2 teraflops? Shouldn't they discard these 2 teraflops if it's only the front-end used to access the IBM system?

  41. mod parent insightful! by Pollardito · · Score: 1

    i'm writing this from my Earth Simulator laptop that will be commercially available next year

    1. Re:mod parent insightful! by Anonymous Coward · · Score: 0

      oooooo oh

      He's telling the truth!! I can tell by the font!!!!

    2. Re:mod parent insightful! by Anonymous Coward · · Score: 0

      Oh... poor kid. I am replying to this from my IBM BlueGene cell phone.

  42. Not very expensive... by Buttonius · · Score: 1

    That lease price is probably per year and a year is approximately 31556909 seconds.
    Assuming a US trillion (1E12), gives 59E12 operations per second, or about 1.86E21 operations per year. That is about 62E12 operations per dollar. There will probably be some (rather significant) additional costs to run and cool the beast...

  43. Re:Ultimate Redudant post by vertinox · · Score: 1

    1. Someone posted the BeoWulf joke 5 posts up.
    2. Supercomputer are basically BeoWulf clusters
    3. Someone else posted the same joke 3 posts up, but included "Will it run linux".
    4. It isn't the first time someone has made the mistake of their/they're
    5. You aren't the first to point that out to anyone.
    6. Somone else will reply to your post and point that out two posts after mine.
    7. Someone else in this comment thread will also make misuse of their and they're.
    8. Somone two posts down from your original will also make a mention of a Beowulf cluster.
    9. And 5 posts after that another will try to get modded funny by trying to include "In Soviet Russia, only old people use Beowulf Clusters Running linux."

    Really... Soviet russia cliches may still be funny, but Beowulf and "Will it run linux" aren't.

    Please think of the children and refrain from mentioning either two ever again. If you are tempted to mention it, it means 5 other people already have by the time somone posts it.

    No one has to say. Really... They don't.

    --
    "I am the king of the Romans, and am superior to rules of grammar!"
    -Sigismund, Holy Roman Emperor (1368-1437)
  44. Mainichi daily news? by anza · · Score: 1

    Isn't that redundant?

    1. Re:Mainichi daily news? by Fry-kun · · Score: 1

      For those who don't speak Japanese, "mainichi" means "daily"

      --
      Did you know that "FTW" ("for the win") is a direct translation of "Sieg Heil"?
  45. yeah but by Trailer+Trash · · Score: 1

    what kind of frame rate does it pull on HL2?

  46. it's filled with Jawas! by illtron · · Score: 1

    Did you see the pictures? It's powered by millions of tiny little Jawas!

    --
    Slashdot: 24 hours behind every other site or your money back!
  47. Re:Ultimate Redudant post by prisoner-of-enigma · · Score: 1

    1. Someone posted the BeoWulf joke 5 posts up.

    That what this all is, a Beowulf cluster of Beowulf cluster jokes. You don't get it, do you?

    2. Supercomputer are basically BeoWulf clusters

    Congratulations! You win the "Duh!" Awards of the day!

    3. Someone else posted the same joke 3 posts up, but included "Will it run linux".

    Humor is in the eye of the beholder.

    4. It isn't the first time someone has made the mistake of their/they're

    This isn't the first time I've corrected them, either. Nor will it be the last. Your ability to use proper grammar, spelling, and punctuation is a reflection on your intelligence level. Ergo, he's an idiot. You're defending an idiot. What does that make you?

    5. You aren't the first to point that out to anyone.

    Yet they never seem to learn...

    6. Somone else will reply to your post and point that out two posts after mine.

    Dept. of Redundancy Dept.

    7. Someone else in this comment thread will also make misuse of their and they're.

    And I'll gleefully point out their flaws as well.

    8. Somone two posts down from your original will also make a mention of a Beowulf cluster.

    See response to #1.

    9. And 5 posts after that another will try to get modded funny by trying to include "In Soviet Russia, only old people use Beowulf Clusters Running linux."

    See response to #3.

    --
    In the end they will lay their freedom at our feet and say to us, Make us your slaves, but feed us. - Fyodor Dostoyevsky
  48. A few minor details by frankie · · Score: 1
    1. It's the fastest in Japan , aka should beat Earth Simulator (40TFlops, ranked #7 last November).
    2. It's basically yet another IBM Blue Gene, but with a much weaker Hitachi attached to it.
    3. #1 ranked BlueGene/L running at Livermore hits 280TFlops.
    4. IBM PPCs dominate the high end of the Top500.
    1. Re:A few minor details by Burz · · Score: 1

      There's an interesting article about the Earth Simulator and climate modeling over here.

  49. Just imagine... by fm6 · · Score: 1

    ...a Beowulf .... AAAACK!!!

  50. Yeah, yeah... by Anonymous Coward · · Score: 0

    As usual, the Japanese press acts like it's a Hitachi product. But really it's an IBM. We know the drill... America sucks, everyone else is great... Japan is so high tech, until you actually go there.

    Hooray!

    What the hell do you do with 59TFlops anyway? Levitation?

  51. Clarification by necro81 · · Score: 1

    This is not, as a misinterpretation of the summary might suggest, the fastest supercomputer on the planet, just the fastest one in Japan. That title of world's fastest is still held by the BlueGene at Lawrence Livermore, which boasts something like 350 teraflops peak. Interestingly enough, this new machine in Japan is a smaller BlueGene computer: same architecture, fewer racks.

    1. Re:Clarification by Anonymous Coward · · Score: 0

      Can you also clarify how fast this supercomputer is in mph, whether the new system sports include tennis, and the exact temperature of the lease price?

  52. How About Some Stats I Can Understand? by airship · · Score: 1

    How many Libraries of Congress will it hold?
    How many football fields does it cover?
    I need some stats I can relate to! :)

    --
    Serving your airship needs since 1995.
    1. Re:How About Some Stats I Can Understand? by scovetta · · Score: 1

      It can run Duke Nukem Forever... in a JVM.

      It's faster than the Quantum Optical Laptop

      And it requires a nuclear power plant to supply the power.

      --
      Wer mit Ungeheuern kämpft, mag zusehn, dass er nicht dabei zum Ungeheuer wird. --Nietzsche
  53. Upside down? by Kontinuum · · Score: 1

    Comparing pictures of this KEK Blue Gene with pictures of Lawrence Livermore's Blue Gene/L, what strikes me is ... one of them appears to be upside-down.

    Any logic to this? Or is the idea that when they add another gazillion processors to each one, they will be able to meet up nicely somewhere over the Pacific.

    1. Re:Upside down? by Anonymous Coward · · Score: 0

      The KEK and LLNL pictures are taken from two different ends. Both of them slant in the same fashion.

    2. Re:Upside down? by PipeIsArt · · Score: 1

      It actually looks like they share the same shape. If you look at the back ends of the IBM Blue Genes pictures you can notice a slight slant where the top seems to extend farther than the bottom. Since you cannot clearly see both ends of either computer, it can seem like they have different shapes.

      --
      I find that although many people are liberal in beliefs, they are conservative in actions.
  54. What happens to them by tie_guy_matt · · Score: 1

    What happens to the "fastest computer yet" after a few years (decades.) They are impressive now but eventually they will become slower than a common desktop machine. Do they have a life cycle where they start as a companie's fastest machine, but later become a server or go into some other duty that doesn't require the world's fastest machine (after the company goes out and buys an even faster machine.) Is there a market for such machines? Or are they so expensive to run that they are eventually just recycled or sold as scrap? It might be cool to have the worlds fastest machine as of 1990 or so to play with. Are such machines still in use?

    1. Re:What happens to them by Anonymous Coward · · Score: 0

      See http://www.cray-cyber.org/ for a sample of some former supers that you can gain public access to.

    2. Re:What happens to them by cr0sh · · Score: 1
      No, the market really isn't there for such machines. There are only a handful of companies and government markets that need this kind of processing power, and they typically need it for very specific reasons: nuclear simulations (bombs and otherwise), molecular modeling simulations (drug companies, mostly), fluid flow dynamics simulations (weather modeling and aerodymanics).

      Many of these needs can be met very well with commodity hardware based cluster machines (which is what this machine sounds like - more or less), with a fat interconnect (Myrinet and others). The old "workhorse" vector-processor supercomputers, like much of the old-skool Cray line - these are machines which are very tough to build, and even tougher to code for. They have their uses, but most "hard problems" are hard because they need parallel processing power - that is, the problems are reducible to simpler problems that can be run in parallel and the solution formed from the aggregate results. Such needs are easily met with the cluster systems - so more of these systems are sold, and the vector boxes have fallen by the wayside (in sales).

      Those places that still need those types of machines tend to horde them like old people horde newspapers. Since these machines are few and far between (even in the heyday of Cray, there really weren't that many of the larger Crays installed worldwide), they all know who has what machines - so if one becomes "decommissioned" at one site, it finds a home at another site as a spare or for parts. That said, the rest of these machines typically become scrap. There aren't many individuals who have the room, let alone the money, to properly store, power, and maintain these boxes. They are big, they are power hungry, and they don't use what one would think of as "standard" parts. Don't even think about the cost in flourinert that some of the Crays requires (gallons upon gallons of flourinert - ever priced the stuff?)...

      Another individual has already posted to you a site that lets the public access a few of these kind of machines if you have the need - such public sites are very rare, but that's OK because so are the machines and the uses for them (I mean, if you need such a machine, and you have a good reason and code for such a need - then you are probably in a handful of people worldwide who would need such access). But yeah - most of them get sold as "scrap". A few land in museums. Others get parted out and land on Ebay (there is one guy on Ebay that has been selling various boards and such parts from a parted-out Cray 3 for a while now)...

      --
      Reason is the Path to God - Anon
  55. Not really faster than Earth simulator by flaming-opus · · Score: 1

    Yes the linpack performance is better, but I think 38tf of Earth simulator is likely to get a lot more real work done than 57tf of blue gene. The blue gene is a really elegant design for creating a very inexpensive supercomputer, which means one can affordably buy a very large system. However, the balance of the system is a little weak in terms of memory bandwidth, and interconnect bandwidth, as compared to the Earth simulator.

    Furthermore, scaling most codes to the tens of thousands of processors of a blue gene is very difficult. Most jobs on the earth simulator, and other large systems, run on a few dozen processors, as the algorithms become communication bound above that number. Algorithm designers are slowly improving this bottleneck, but it's something that must be done anew for each piece of code.

    Not to say that this isn't a fantastic machine, and one offered at a very modest price for the capability. I simply object to the title of "Japan's fastest supercomputer".

    1. Re:Not really faster than Earth simulator by weffew... · · Score: 1

      Having been involved with BG/L I'd have to disagree with your reasoning, though I fundamentally agree with your point.
      The toriodal network/barrier net that IBM Research invented (and implemented-on die) for the BG/L is actually quite impressive, and gives quite enough bandwith for the puny processors (especially given the wormhole routing implemented in hardware).

      Where the BG/L design falls down somewhat is that the individual nodes have only a limited amount of RAM.
      It's not so good for applications with large datasets (eg weather modelling).
      Wef

  56. Designing drugs by Anonymous Coward · · Score: 0

    Designing better drugs by simulating molecular interactions and modelling virus, bacterial, cancer mechanisms. Duh.

    That said, the politicians will probably use it for something useless as usual.

  57. slightly confused why... by weffew... · · Score: 1

    ...why the Hitachi (Japanese) tiddler (~2TFlops) gets the same billing as the IBM machine (~57TFlops, built in Rochester, USA).

    Oh wait - when added together, these two make a *Japanese* machine.

    Forgive me, but if >90% of the complexity comes from the USA they might as well not have bought the Hitachi system in the first place.

    This is just a national pride/publicity thing in my opinion - the Japanese establishment want to have a machine to crow over.

    Wef

    1. Re:slightly confused why... by Anonymous Coward · · Score: 0

      Japan's previous fastest supercomputer was entirely of Japanese origin (it was made by NEC) and it is considerably faster than the 'small' Hitachi supercomputer, so there goes the national pride theory. I suspect this combination was chosen because the IBM machine excels in different operations than the Hitachi does.

  58. 59 terra flops by RMB2 · · Score: 1

    IANAE of supercomputers by any means, but to me it seems that the first 57.3 terra flops makes adding the other 2.15 terra flops rather superfluous. Isn't that like DuctTaping an Estes-C onto a Saturn V?

    --
    [/sarcasm]
    1. Re:59 terra flops by Anonymous Coward · · Score: 0
      Isn't that like DuctTaping an Estes-C onto a Saturn V?

      Well like the post above you pointed out, it'd be more like strapping an Estes-C onto a Saturn V then calling it an Estes rocket...

  59. Supercomputer Accessories by camusflage · · Score: 1

    I hear the NOS sticker is good for two teraflops, and if they put a big unpainted wing on top of it, that's another three teraflops. Stage III, including the watermelon shooter con exhaust pipe, takes it up an additional ten teraflops!

    --
    The truth about Scientology, Xenu, and you: Operation Clambake
    1. Re:Supercomputer Accessories by RMB2 · · Score: 1

      That's a little concept we like to call "visual horsepower", as opposed to "actual horsepower".... seems only logical that the same thing should apply to supercomputers

      --
      [/sarcasm]
  60. Dull boring boxes by Antony+T+Curtis · · Score: 1

    Is it me... or is it that ever since Seymore Cray died, supercomputers decided to take on a dull filing-cabinet look rather than the futuristic look?

    Ok, IBM gets some points for the "its not a cube, it slanted" look... but where are the blinky lights? Surely for as much money that is paid, a few pennies can be diverted to the visual appeal?

    --
    No sig. Move along - nothing to see here.
    1. Re:Dull boring boxes by cr0sh · · Score: 1

      Do some googling on "CAM Brain Machine" if you want to see a very interesting (from all angles) computer design. Not only was this computer's physical design something to marvel over, but the idea behind how it worked (whether it worked well or not is another story) was pretty interesting, too. Think "on-the-fly evolutionary neural net processing", done in hardware using a specific (now discontinued) model of Xilinx FPGA. I would so love to pick one of those "boxes" up off of Ebay (probably have an easier time buying a Cray 3 from there, though)...

      --
      Reason is the Path to God - Anon
    2. Re:Dull boring boxes by Anonymous Coward · · Score: 0

      What do you mean? http://www.cray.com/ is very much alive....

  61. Rumor has it... by symbolic · · Score: 1

    ...that it's just fast enough to keep up with Microsoft's upcoming Vista OS.

  62. Duke! by Anonymous Coward · · Score: 0

    It may be able to run Windows... but it sure as hell won't be able to run DNF!

  63. Sound Familiar by Nom+du+Keyboard · · Score: 1

    Sounds a lot like Intel's first dual processors. Let's just slap together a Hitachi and an IBM, add up the TFlops, and claim victory.

    --
    "It's the height of ridiculousness to say for those 9 lines you get hundreds of millions."
  64. suggestion by Zork+the+Almighty · · Score: 1

    FLOPS/(W^2*cm^3) ?

    --

    In Soviet America the banks rob you!
  65. The silver lining... by Saeger · · Score: 1

    ...is that with fewer /. comments (+4) to obsessively read through, I find I suddenly have more time to get other stuff done. I, for one, hope that this less-cream-filters-to-the-top 'feature' stays. :)

    --
    Power to the Peaceful
  66. Long time for AI, never for "god-like" intelligenc by raftpeople · · Score: 1

    Earliest estimates for simulating 1 human brain in real time using a very basic neural network setup is somewhere around the year 2020. This is just based on number of neurons and number of connections. When you throw in the glial cells that were previously thought to be relatively unimportant, despite being a large percentage of the brain, then it's probably further still than 2020.

    Additionally, to try to simulate ALL of the chemical and electrical interactions/processes that go on inside the brain (just to be sure there isn't something important in those details), then it gets pushed out even further.

    So, 1 human brain is a long way off, and that doesn't even take into account making the thing do something useful, it's just the raw processing power. To achieve AI requires understanding of how the brain works, so push it of even farther (further?).

    Finally, intelligence is purely a mapping between specific goals, and the actions required to achieve those goals (problem solving by abstracting prior experience and pattern matching to current conditions). There is a mis-conception that you could have "god-like" intelligence that is all knowing and can even give the "correct" answer always, etc. The answer to any problem is specific to the conditions that surround it. Although I don't have a mathematical proof that says it's impossible to have this perfect "god-like" intelligence, I strongly suspect that it is mathematically provable.

  67. A Sandcrawler? by Bnderan · · Score: 1

    It looks like it's full of Jawas.

    1. Re:A Sandcrawler? by roboteeruk · · Score: 1

      That's exactly what I thought when I saw it. I was expecting some minature jawas to be milling around it.

  68. Yeah, but.. by tpemble · · Score: 1

    Does it run OSx86?

  69. Ter(r)a by msbsod · · Score: 1

    Terra is also the Latin word for earth. Tera stands for the factor 10^12.

  70. Re:Ultimate Redudant post by Retric · · Score: 1

    Your ability to use proper grammar, spelling, and punctuation is a reflection on your intelligence level.

    No, your ability to use proper grammar, spelling, and punctuation is a reflection of your education not just your intelligence level. Try talking with a few people outside the US, which use English as a 3rd language, and you might learn how meaningless your supposed link is.

  71. Imagine a beowulf cluster... by TechnoGuyRob · · Score: 1

    Sorry, I just had to do that.

  72. [OT] Mainichi Shinbun by minus_273 · · Score: 1

    Any other Japanese and English speakers out there find Mainichi Daily news to be a little redundant? :-p

    --
    The war with islam is a war on the beast
    The war on terror is a war for peace
  73. Re:Long time for AI, never for "god-like" intellig by Just+Some+Guy · · Score: 1
    When you throw in the glial cells that were previously thought to be relatively unimportant, despite being a large percentage of the brain, then it's probably further still than 2020.

    Not by much, though. Ray Kurzweil makes a good case that the price-performance of computers has been doubling in just over a year, and that the rate of change itself is increasing. So, if simulating glial cells require 100 times the computing power of simulating only neurons, then it should be possible in less than seven years after the first neural simulator would be possible.

    Although I don't have a mathematical proof that says it's impossible to have this perfect "god-like" intelligence, I strongly suspect that it is mathematically provable.

    While I think you're probably right on that point, I also think that's a straw man. If we're capable of building a computer with 1,000 times the human-style intelligence of a biological person, then sure that would be of interest - even if it weren't actually infinitely smarter than we are. Singularity advocates contend that we only need to build a computer smarter than the best computer designers, and then step back to let the positive feedback loop take care of the rest. I truly believe this will happen in my lifetime, and probably before I turn 50.

    --
    Dewey, what part of this looks like authorities should be involved?
  74. Good thing it's fast by ThinkWeak · · Score: 1

    ... cause they only have it for 5 years

  75. Sure thing but... by whitepony02027 · · Score: 0

    Sure its fast and all of that but does it play WoW?

  76. Re:Long time for AI, never for "god-like" intellig by raftpeople · · Score: 1

    ...Ray Kurzweil makes a good case...

    I have not read Ray's book, but the recap I did read made me think it was not worth reading. I don't remember all the details other than I logged him away in my brain as a person that was not making valid claims. I could be wrong.

    While I think you're probably right on that point, I also think that's a straw man. If we're capable of building a computer with 1,000 times the human-style intelligence of a biological person, then sure that would be of interest - even if it weren't actually infinitely smarter than we are. Singularity advocates contend that we only need to build a computer smarter than the best computer designers, and then step back to let the positive feedback loop take care of the rest. I truly believe this will happen in my lifetime, and probably before I turn 50.

    I agree it is very interesting and exciting to achieve almost any level of success in this field, especially a 1,000-times-smarter-than-humans computer. But my gut tells me that goals/motivation are as important a topic in this brain as the pure processing power. How do you get it to do what you want? I think we will make steady progress, I just think there is far more work than many realize.

  77. Re:No project to run on it by madmarcel · · Score: 2, Interesting

    > They're gonna ask the public for research themes? ... AFTER THEY BOUGHT IT???

    This is not unusual.

    Universities around the world are filled with expensive equipment that doesn't get much use.
    What usually happens is that somebody has a vague idea for a research project, applies for a funding grant, but doesn't expect to get it, and then three years down the track the grant gets approved, and you have to buy something that you never expected to get, and you're not sure what to do with it.

    Part of my job is making such expensive purchases accessible by as many people as possible.
    I have access to 4 clusters at the moment, and all four have been 99% idle/underused from the day they were bought, and will continue to be so until I've had a go at them. I've got one cluster to the point that it is now being used to 100% of it's capacity and I've just started on the second one. (This second one has been mostly idle for nearly two years now)

    It's quite interesting to note that once people can make full use of a cluster, they will quickly come up with ideas and projects that will start pushing the cluster to it's full capacity.

    Heck, Weta FX (Responsible for in Lord of the Rings and King Kong, but I'm pretty sure you knew that ;)
    ran short of rendering grunt for the final battle scene in Return of the King.
    (The battlescenes with all the oliphaunts)
    So they rang the producers and got funding for another IBM Blade cluster, at approx $4m (something like that)
    Bought it, build it, used it for LOTR, used it for a little bit of rendering for two other movies - Zorro and Van Helsink I think - and then it sat idle for a year!

    I'll start work on that one soon, I've got access about 150+ 2.8Ghz Intel Xeons that need a good workout ;)

    (No, none of the clusters run Windows - sorry - and not a single beowulf cluster ;^)

  78. Re:Long time for AI, never for "god-like" intellig by Just+Some+Guy · · Score: 1
    I have not read Ray's book, but the recap I did read made me think it was not worth reading.

    I'm thoroughly enjoying it. What impressed me the most is that it has less than 500 pages of content, then 200 pages of notes and references. I don't know whether his predictions will be correct, but he certainly makes the raw data that he analyzed available for you to draw your own conclusions.

    How do you get it to do what you want?

    Well, that's definitely the scary part. It seems quite likely that an entity much smarter than a person might have goals and motivations that are perfectly reasonable - and completely alien.

    I think his "weaker" proposition, that we eventually become the AI as our biological neurons are gradually replaced by nanotech work-alikes to the point that we're predominantly nonbiological, seems more likely in the "short" term.

    --
    Dewey, what part of this looks like authorities should be involved?
  79. Yes, but by Doug+Neal · · Score: 1

    Does it use AJAX?

  80. yes, but... by advocate_one · · Score: 1

    can it run Vista?

    --
    Donald 'Duck' Dunn: We had a band powerful enough to turn goat piss into gasoline.
  81. Daily Daily news? by jwiegley · · Score: 1

    I don't care about the story. I just thought the redundancy of the newspaper's translated name was rather humorous.

    --
    I will never live for sake of another man, nor ask another man to live for mine.
  82. Yes but... by MarkVVV · · Score: 1

    Does it run linux?

  83. Micropayment options? by WormholeFiend · · Score: 1

    Do they offer micropayment options for really small calculations?

  84. Comment removed by account_deleted · · Score: 1

    Comment removed based on user account deletion

  85. Yeah but... by Ironballs · · Score: 0

    ... how many people can you put in a conference call using Skype?

  86. Re:Ultimate Redudant post by prisoner-of-enigma · · Score: 1

    No, your ability to use proper grammar, spelling, and punctuation is a reflection of your education not just your intelligence level. Try talking with a few people outside the US, which use English as a 3rd language, and you might learn how meaningless your supposed link is.

    As a measure of respect for the language you're speaking -- whatever language that may be -- the least you can do is speak it (or write it) correctly. If I were speaking in German, criticism of improper grammar would be equally valid. This criticism is not restricted to Anglo-centric languages despite your attempt to paint it as such.

    If you can't do it right, don't bother doing it at all. In this day and age of grammar checkers, spell checkers, and the resources of the Internet in general, no one can use the excuse that they didn't know better. To do any less excuses laziness and invites mediocrity. Typographical errors are one thing, but the obvious grammatical inaccuracy shown in the OP has no excuse whatsoever.

    --
    In the end they will lay their freedom at our feet and say to us, Make us your slaves, but feed us. - Fyodor Dostoyevsky