Slashdot Mirror


Inside The World's Most Advanced Computer

Junky191 writes: "Just came across an informational page for the Earth Simulator computer, which provides nice graphics of the layout of the machine and its support structure, as well as details about exactly what types of problems it solves. Fascinating for the engineering problems tackled- how would you organize a 5,120 processor system capable of 40Tflops, and of course don't forget about the 10TB of shared memory." Take note -- donour writes: "well, the new list of supercomputer rankings is up today. I have to say that the Earth Simulator is quite impressive, from both a performance and architectural standpoint."

270 comments

  1. AC's Are on Your ASSSSSSSSSS!!!!!!! by Anonymous Coward · · Score: 0, Informative


    Biotches! I'll 0wn the CLIT soon

  2. Apples and Oranges by Anonymous Coward · · Score: 0

    Isn't it fairly difficult these days in terms of the specilisation that occurs at this end of the market to really compare between different machines?
    Surely purpose is the best way to organise the rankings?

    1. Re:Apples and Oranges by Jeremy+Erwin · · Score: 3, Interesting

      Specialization? What specialization? The Top500 rankings are based on LinPack-- a software package for solving dense systems of linear equations-- which seems applicable to a fairly general set of scientific problems.

      Do supercomputing manufacturers cheat on benchmarks? I don't know. Presumably it would be a rather expensive proposition-- and since supercomputing sites will benchmark with a variety of specialized and general purpose libraries, it seems unlikely to work.

      There, are, of course, differences between weather simulations and galactic evolution simulations. But field specific benchmarks are inappropriate for a site like Top500--the whole point of the site is to allow someone to analyse gross trends. "This memory architecture once dominated the rankings--now its used by only a few entries. Perhaps our next computer platform shouldn't be based on that architecture." (and possibly writing journal articles about it.)

      In addition, general purpose supercomputing sites are relatively common.

    2. Re:Apples and Oranges by Anonymous Coward · · Score: 0

      oh yeah! the americans aren't winning, the japanese must have spent a huge amount of money on this specialised linpack supercomputer (which somehow cheats). obviously the american one is much faster for doing anything useful.

    3. Re:Apples and Oranges by cperciva · · Score: 2

      The Top500 rankings are based on LinPack-- a software package for solving dense systems of linear equations

      Almost, but not quite. The Top500 rankings are based on solving the dense system of linear equations generated by the LinPack benchmark driver. Using LinPack is optional. For that matter, using Gaussian elimination with partial pivoting is optional, as long as you meet the error bound of O(n \epsilon).

  3. Earth Simulator? by Artifice_Eternity · · Score: 5, Funny

    Didn't the mice in H2G2 already build such a computer? I think it was called... the Earth.

    Will the Earth Simulator have the nice fjords by Slartibartfast? :)

  4. Uhhhhhhhh... by Anonymous Coward · · Score: 1, Funny

    Imagine a beo*SLAP*

  5. bold name by SlugLord · · Score: 1, Interesting

    "Earth Simulator" is a rather bold name for a supercomputer, especially when you consider it probably can't even simulate the global weather fast enough to predict it (or even tell you what the weather is in real time). The computer looks impressive, but I think they should have stuck to a more abstract name rather than what I see as false advertisement.

    1. Re:bold name by Anonymous Coward · · Score: 0

      " but I think they should have stuck to a more abstract name rather than what I see as false advertisement."

      Yea because we all know that all the other companies only use product names which describe their products capabilities exactly. When you pushing that much processing power around you can dam well call it whatever you want! Did you even look at the numbers?

    2. Re:bold name by Anonymous Coward · · Score: 0

      I think SlugLord would have prefered an abstract name like:

      Hyper eForce SR27000/MPP+ SX-Turbo-BigAssCoreDumps

    3. Re:bold name by __aaklbk2114 · · Score: 2, Funny

      What about: Super ASCI White Turbo Champions Special Edition

    4. Re:bold name by Anonymous Coward · · Score: 0

      Yeah thats just because you poofter americans are jealous that another country has come out and blown you all off the map. hey, sorry but China could too.

    5. Re:bold name by netsharc · · Score: 1

      I wonder if the computer simulates the effect on local temperature itself has.

      --
      What time is it/will be over there? Check with my iPhone app!
    6. Re:bold name by fb · · Score: 1

      >especially when you consider it probably can't
      >even simulate the global weather fast enough to
      >predict it

      The global weather can be - and is, on a daily basis - easily predicted by much slower computers. This one was built for *climate*, not weather simulations. While attending the NEC User Group conference I made some photo of the E.S. complex myself which I might or not put online at my site (see profile).

      The early benchmarks on real applications are quite impressive.

      Anyway these are old news. I submitted an early story about the Earth Simulator one year and a half ago, when it was being built, and it was rejected. Go figure.

      --
      fB
    7. Re:bold name by Arcturax · · Score: 1

      It's a SIMULATOR. It won't and probably can't be perfect. But it does get us closer!

      I suppose Maxis should be sued for the Sims and Sim City and Sim Earth as well because they aren't perfect? Please.

      --

      --Won't that be grand? Computers and the programs will start thinking and the people will stop. - Dr. Walter Gibbs
    8. Re:bold name by Anonymous Coward · · Score: 0

      The Erth Simulator has already predicted weather patters quite well. If you look on their site, they have predicted hurricanes (I read the page awhile back) that were very very close to the path of the hurricane that took place soon afterwards.

    9. Re:bold name by jstott · · Score: 1

      Earth Simulator" is a rather bold name for a supercomputer, especially when you consider it probably can't even simulate the global weather fast enough to predict it (or even tell you what the weather is in real time).


      The computer isn't intended to forcats the weather. It's for forcasting the climate. Weather is a local and worries about local day-to-day fluctuation. Climate deals with large-scale (continent-sized) regions and how the averages vary on a scale of decades to centuries.


      The other intended use, as others have already pointed out, is to validate climate models against historical data and to optimize non-linear model parameters.


      -JS

      --
      Vanity of vanities, all is vanity...
  6. I'd like to learn how to code for this beast by Anonymous Coward · · Score: 3, Funny

    So that in 15 years I'll already know how to code for the PlayStation 6.

    1. Re:I'd like to learn how to code for this beast by Anonymous Coward · · Score: 0

      Playstation 6? I've got an old Altair homebrew you can practice on, if that's the case.

  7. Not a Beowulf cluster comment, but... by rob-fu · · Score: 0, Interesting

    if you could install the clients on this thing, you could find the cure to cancer, crack RC5-64 and OGR-25, decipher all of the SETI@Home work units (but you still wouldn't find any aliens :)...hell, you could solve ALL of the distributed computing applications on this machine.

    Stating the obvious, that's a shitload of CPU power. :)

    1. Re:Not a Beowulf cluster comment, but... by Anonymous Coward · · Score: 0

      yeah, but will it play games man!?!? Who the hell cares about modelling the earth when what we really need is a really big multi-user MMORPG to play. Finally this may be the machine big enought to finally do the job right!

    2. Re:Not a Beowulf cluster comment, but... by sacremon · · Score: 4, Interesting

      Better rethink that theory:

      Earth Simulator: 35.86TFlops.sec (according to Top100 list)

      Seti@Home network: 37.07TFlops/sec (over last 24hr., according to the site).

      Just because it is an incredibly powerful machine doesn't mean it has the distributed computing projects beat.

      --
      If you can't beat them, embrace and extend them.
    3. Re:Not a Beowulf cluster comment, but... by sgtsanity · · Score: 1

      Speaking of cracking those codes, the NSA would love to get their hands on this, and I wouldn't be suprised if they're either leasing some time on it or have something up their sleeves just as powerful.

    4. Re:Not a Beowulf cluster comment, but... by NeoSkandranon · · Score: 2, Insightful

      Flops aren't everything. Until the SETI@home network has the bandwith that the Earth Sim has between processors (someone said 1Tb/sec?) until then one supercomputer will have it beat for anything that requires data to be compared (has this chunk of data been processed by one of the other 4999 processors yet?)

      --
      If you can't see the value in jet powered ants you should turn in your nerd card. - Dunbal (464142)
    5. Re:Not a Beowulf cluster comment, but... by Leperflesh · · Score: 2, Insightful

      So if they were to install Seti@Home on the earth simulator, when it was not busy simulating Earth, Seti@home's speed would double...

      -Leperflesh

      --
      I am allowed to criticize you: you are not allowed to criticize me. Sorry, that's just how things are.
  8. base 10 measurement of memory by SlugLord · · Score: 2, Insightful

    I guess this is a *little* off-topic, but this really bugs me. They're building this really cool supercomputer, and they list the memory with base-10 prefixes instead of the standard base-2. I mean I can almost understand when dell does that with hard drives (it pumps up the number for advertising purposes), but it's just silly in a scientific arena.

    1. Re:base 10 measurement of memory by Anonymous Coward · · Score: 0

      It really bugs me that people have come to think that base 2 is, or should be, standard. The whole point of the SI is to standardize the meanings of measurmements. All those G-, k-, M- prefixes are from the SI. The SI is base 10.

      Base 2 is easier for computers, and certainly a de-facto standard. But it undermines the need for a consistent measurement system. It's almost as bad we stubborn Americans who take pride in our idiotic system of measurement.

    2. Re:base 10 measurement of memory by mt-biker · · Score: 0

      The fact is that "kilo-" means 1000, "mega-", 1000000, and "giga-" 1000000000. They're defined that way. _That's_ a standard.

      You want to explain why base-10 figures shouldn't be used. Other than calling it "silly", and blaming it in marketing, that is?

    3. Re:base 10 measurement of memory by Anonymous Coward · · Score: 0

      Its silly because the computer doesnt measure its specs in decimal. And you are interested in what the computer can use. The sometimes confusing use of kilo, mega and so fort in computer specs should still be exact to the level computer works

      The user problems arise with all the level of detail that 1024 represent instead of using kilo and the fact that the difference between a "round" number computer wise and a round number decimal, soon becomes significant.

      So the adaption og kilo,mega and so on, was a simple user friendly solution to a growing problem. Its intended audience understood the problem and why it was a good solution, and the rest doesnt matter... the details around the problem covers the details over the solution.

      If we need to keep consistency to man made standards for general human understanding , we better start developing decimal computers.

  9. Hmmm.. by Peridriga · · Score: 4, Funny

    Could you imagine a beouwolf..... Ahh fuck it..

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

      Thankyou for generating Slashdot autoresponse #83

  10. OS'es for the supercomputers... by dnaumov · · Score: 4, Interesting

    I am not going to ask "Does this run Linux ?" because it obviously does not, but can anyone point to some good resources on what kind of Operating Systems do these monster machines run ? Are they some kind of a UNIX ? Or are they some elite breed of OS that mortal humans have no chance of understanding ? Linkage appreciated.

    1. Re:OS'es for the supercomputers... by Anonymous Coward · · Score: 0

      DOS

    2. Re:OS'es for the supercomputers... by rob-fu · · Score: 1

      Here is a link for ASCI Red @ Sandia National Labs.

      From the article:
      The system uses two operating systems to make the computer both familiar to the user (UNIX) and non-intrusive for the scalable application (Cougar). And it makes use of Commercial Commodity Off The Shelf (CCOTS) technology to maintain affordability.

      Hmm, I see one familiar OS in there...

    3. Re:OS'es for the supercomputers... by binaryDigit · · Score: 3, Funny

      The IBM units run AIX, the DEC^H^H^HCompaq^H^H^H^H^H^HHP systems run Tru64 Unix, which I believe is derived (or simply renamed) Ultrix. Don't know what ES runs though, maybe extended mode DOS 6.2?

    4. Re:OS'es for the supercomputers... by Elbereth · · Score: 2

      You know, that "^H" thing was funny the first quadrillion times I saw it, but since I saw your post, it has suddenly stopped being amusing. Might I suggest ^? for your next post? I'm quite sure I'll find that amusing for the next quadrillion times I see it.

    5. Re:OS'es for the supercomputers... by binaryDigit · · Score: 2

      Sorry, but belive it or not, I've yet to see anyone do it. Didn't realize it was already hackneyed, though I guess I should have known since it was kinda an obvious thing and all. How about Digital Compackard Corp?

    6. Re:OS'es for the supercomputers... by vespazzari · · Score: 1

      could someone explain the ^H thing? Im not getting it. thanx

      --
      "Alcohol, cause of, and solution to, all of life's problems" -Homer Simpson
    7. Re:OS'es for the supercomputers... by Anonymous Coward · · Score: 0

      silly boy^H^H^H^H^H^H^H^H^Hsilly boy

    8. Re:OS'es for the supercomputers... by binaryDigit · · Score: 2

      Ahhh young man, sit down and let me explain to you how when I was your age I had to walk 20 miles through the snow uphill to be able to use a pc (my apologies if you're not a young man :)

      Whenever you see the caret (^) followed by a letter, it typically means CTRL+letter. So ^H would mean that holding CTRL while pressing the H. These are ASCII codes that used to do various things on terminals. In the case of CTRL-H, it would be interpreted as BackSpace. ^G is another common one and causes your BELl to ring. Try it from your command line (even works in Windoze). Look up an ASCII chart for those values less than 32 to see what they do.

    9. Re:OS'es for the supercomputers... by __aaklbk2114 · · Score: 0

      It's running Windows XP Embedded with IE removed so it will boot up faster. Duh!

    10. Re:OS'es for the supercomputers... by heydrick · · Score: 2, Informative
      NEC machines like the Earth Simulator run a version of UNIX called SUPER-UX or S-UX, for short.

      NEC Press release mentions SUPER-UX.
      NEC SX-6 page has lots of info.

    11. Re:OS'es for the supercomputers... by mt-biker · · Score: 4, Informative

      These machines tend to be clusters of smaller machines. IBM's SP architecture, for example, runs AIX which doesn't need to scale particularly well.

      The magic in SP is partly hardware (high-speed interconnect between nodes), partly the admin software which allows admin tasks to be run simultaneously of many nodes (a non-negligible task), and is otherwise left up to the application programmers to use MPI or similar to get the application to run over the cluster.

      Single system images typically don't scale this large. Cray's UNICOS/mk (Unix variant) is a microkernel version of the UNICOS OS, used on the T3E and it's predecessors, where a microkernel runs on each node, obviously incurring some overhead, but avoiding bottlenecks that otherwise occur as you scale. Here's some info. Last time I checked, T3E scaled to 2048 processors.

      Out of the box, SGI's IRIX scales very nicely up to 128-256 processors. Beyond that "IRIX XXL" is used (up to 1024 processors, to date). This is no longer considered to be a general purpose OS!

      IRIX replicates kernel text across nodes for speed, and kernel structures are allocated locally wherever possible. But getting write access to global kernel structures (some performance counters, for example) becomes a bottle-neck as the system scales.

      IRIX XXL works around these bottle-necks, presumably sacrificing some features in the process. Sorry, I can't find a good link on IRIX scalability.

    12. Re:OS'es for the supercomputers... by Surak · · Score: 2

      Don't know what ES runs though, maybe extended mode DOS 6.2?

      "Non--system disk or disk error. Replace disk and
      "Strike any key to continue . . ."

      Shit! Where's the floppy drive on this thing! :)

    13. Re:OS'es for the supercomputers... by Anonymous Coward · · Score: 0

      You really ought to consider using ^W ;)

    14. Re:OS'es for the supercomputers... by Paul+Jakma · · Score: 2

      Ultrix was a BSD derived Unix.

      Tru64 is nee Digital Unix nee OSF/1 - a project that came about when DEC, HP and IBM came together to found The Open Software Foundation (OSF) to develop "open" unix (they felt threatened by Sun i think). The OSF released OSF/1, a System V R2 based Unix, which was adopted by DEC as it's new Unix to replace RISC Ultrix.

      --
      I use Friend/Foe + mod-point modifiers as a karma/reputation system.
    15. Re:OS'es for the supercomputers... by foobar104 · · Score: 2

      Out of the box, SGI's IRIX scales very nicely up to 128-256 processors. Beyond that "IRIX XXL" is used (up to 1024 processors, to date).

      Do you have any information to back this up? I don't work for SGI, but I work closely with them, and I've never heard the term "IRIX XXL." I've worked on the 768-processor O3000 system in Eagan, and as far as I noticed it was just running stock IRIX 6.5.14 (at that time).

      Then again, I've never used a 1-kiloprocessor system, either. So maybe we're both right.

    16. Re:OS'es for the supercomputers... by Codifex+Maximus · · Score: 2

      Abort, Retry, Fail?

      Keyboard Error - Press any key to continue...

      Error 0 - There is no message for this error.

      Current Disk is no longer valid!

      Anyway, I'll bet the ES runs Plan9 or Hurd.

      --
      Codifex Maximus ~ In search of... a shorter sig.
    17. Re:OS'es for the supercomputers... by Surak · · Score: 3, Informative

      Nope. As others have pointed out it runs Super-UX, which is a commercial Unix variant all NEC's supercomputers run.

      Anyway, it probably S-UX. :)

    18. Re:OS'es for the supercomputers... by Anonymous Coward · · Score: 0

      It seems that there are 2 versions of IRIX 6.5, AWE for workstations and ASE for servers. They only differ in the extra software they come with not the core OS.

      SGI offers preconfigured single-image systems with up to 512 CPUs running standard IRIX. Above 512 it looks like you need to sit down and talk with them, though I've never seen any info about OS differences between the "small" and "large" machines. Isn't there an SGI Origin installation @ SNL that has 2048 CPUs?

      Some links:
      http://www.sgi.com/software/irix6.5/datash eet.pdf
      http://www.sgi.com/software/irix6.5/techb rief.pdf
      http://www.sgi.com/origin/3000/

    19. Re:OS'es for the supercomputers... by sys.napalm · · Score: 2, Informative

      The SGI Origin2000 scales up to 512 processors. XXL refers to the product ID for IRIX for the larger versions of the O2k. SGI Origin3000 scales up to 1024 processors but is only sold as an actual product up to 512 processors. SGI has sold O3k's with over 512 cpus in a single system image to some customers, such as NASA, but it's not treated as an actual product.
      Standard IRIX therefor scales up to 256 processors on O2k and 512 processors with the XXL version. The only difference between the two is that drivers for the former might not work with the later because a few kernel structures changed. The same is true for the Origin3000 versions of IRIX.

    20. Re:OS'es for the supercomputers... by Anonymous Coward · · Score: 0

      Anything beyond 128P uses XXL. The XXL kernel has been intergrated into the standard IRIX release.

  11. the meaning of life by Axeus · · Score: 0

    I thought the Earth was a supercomputer?
    Btw, the the meaning of life is 42.

    1. Re:the meaning of life by Oily+Tuna · · Score: 1

      a) The HHGTTG is fiction
      b) Just because something is a [fictional] supercomputer does not preclude the existance of another, superer computer simulating it.

      --
      Mmmmmmm ... sushi.
    2. Re:the meaning of life by Anonymous Coward · · Score: 0

      Why do people constantly quote this stupid book in every damn slashdot story? Yeah, it was funny and I read it two or three times over the years. But let it go, man...

    3. Re:the meaning of life by snoozebutton · · Score: 0

      NEVER! ;)

    4. Re:the meaning of life by caveat · · Score: 1

      yeah, the earth is a supercomputer...the perfect analog simulation machine of the earth. same goes for the universe, as my physics/cosmology professor used to say. doesn't mean we can sit down at the console though.

      --

      Facts do not cease to exist because they are ignored. - Aldous Huxley
  12. Just came "ACROOS"? by reynolds_john · · Score: 0, Offtopic

    You don't say.

  13. Why by alphaCoward · · Score: 1

    What does: "It is now under development aiming to start from FY2001." mean? am i missing something here...

    1. Re:Why by nr · · Score: 1

      Dunno, but FY is used as short for "Fiscal Year".

  14. I have a dream by perthstyle · · Score: 0, Informative

    ..my friend called Richard will actually take his cock out of his own ass one day! Hi Nanna!

  15. Where would the SETI project rank? by Anonymous Coward · · Score: 0

    If we counted the average speed of all the SETI volunteers over time, does anyone have an idea where it would rank?

    1. Re:Where would the SETI project rank? by willpost · · Score: 2, Informative

      "How does the computing power of Seti@home compare with existing supercomputers?"
      "The most powerful computer, IBM's ASCI White, is rated at 12 TeraFLOPS and costs $110 million. SETI@home currently gets about 15 TeraFLOPs and has cost $500K so far"

      Earth Simulator Project Total peak performance: 40 Tera FLOPS.

      Of course the systems' architecture is different so using the speed to evaluate processing power is difficult.

      There's a TFlop chart on the earth development button.

      Simulating the Earth down to square kilometers will be impressive.

    2. Re:Where would the SETI project rank? by Anonymous Coward · · Score: 0

      http://setiathome.ssl.berkeley.edu/totals.html
      Th e seti@home project is on 37.07 TeraFLOPs/sec, and this is NOT peak performance.
      So the fastest computer on earth isn't faster than the combined effort of nerds around the globe.
      But offcourse the earth computer doesn't need to run in parallel... Eat my backbone!

  16. Eniac... by chuckw · · Score: 3, Interesting

    What is really amazing is that in 50-60 years, this amount of computing power will easily fit within the confines of the standard PC case (assuming such a thing even exists 50-60 years from now). Remember ENIAC...

    --
    *Condense fact from the vapor of nuance*
    1. Re:Eniac... by Mizery+De+Aria · · Score: 0

      I predict it'll be commonplace before 2010.

      --
      If you're religishitty, KILL YOURSELF!
    2. Re:Eniac... by Anonymous Coward · · Score: 2, Interesting

      Where did you get that 50-60 years figure from? Today's PCs can do about 0.5 Gflop/s. Assuming an 18-month Moore's Law, we'll get 40 Tflop/s in our PC case in...

      1.5*log(40000/0.5)/log(2) = 24 years.

    3. Re:Eniac... by Anonymous Coward · · Score: 0

      No such thing as a Moore law exists. It's just an observation without any theoretical support. On the other side we see that miniaturization bumps into quantum effects and the computation speed is seriously limited by the speed of light already. So I think the only way to speed up more is just add more processing elements which is easily done but requires much space.

    4. Re:Eniac... by Surak · · Score: 2

      If Moore's law doesn't exist, how come it's remained fairly constant at least since the introduction of the Intel 4004? Your "miniaturization bumping into quantum effects" argument was used back in the days of the 486. Many industry analysts said at the time that we'd never get faster than a 486/66. Then they said we'd never get faster Pentium at about 500--600mhz. And so on.

      They're getting around the miniaturization bumps with new manufacturing processes and new materials. Copper technology was the last breakthrough, who knows what the next one will be?

    5. Re:Eniac... by discstickers · · Score: 1

      Uh... I'm pretty sure Apple has a machine that can do more than .5GFlops. And Gflops/s is kinda redundant, no?

      (ps, ps means per second =)

      --
      I have a shitty sig!
    6. Re:Eniac... by justinstreufert · · Score: 1

      ps might mean "per second" or it also might be the last two letters of "ops," signifying "operations" ;)

      Justin

      --
      "Why would God give us a waist if we wasn't supposed to rest our pants on it?" - Rev. Roy McDaniels
    7. Re:Eniac... by Anonymous Coward · · Score: 0

      And you might also care to note he wrote Gflop/s...

  17. the real question is... by doooras · · Score: 4, Funny

    how many FPS does this bitch get in Quake III?

    1. Re:the real question is... by Anonymous Coward · · Score: 0

      What is the refresh rate of its monitor?

    2. Re:the real question is... by Anonymous Coward · · Score: 0

      Actually most of these would be slower than my cheap -o PC. They are usually designed to do a lot of stuff quickly rather than 1 thing really fast. So they could run 10,000 instances of Quake, although the question why would come up, but one instance is unlikely to run much faster, than 100 FPS at 1600x1200.

    3. Re:the real question is... by Lars+T. · · Score: 5, Funny

      Does it have a floppy drive?

      --

      Lars T.

      To the guy who modded me down from perfect to terrible Karma - Apple haters still suck

    4. Re:the real question is... by Tony+Hoyle · · Score: 2

      Probably not many. Computers of this size generally don't have exceptionally fast processors, just shitloads of them. Since Quake is a (mostly) single threaded app a standard PC would run it faster.

      Then again you could write a massively parallel i686 emulator (precalculating 5,120 instructions simultaneously) and run the worlds fastest (and most expensive) PC.

    5. Re:the real question is... by Toraz+Chryx · · Score: 2

      Someone install BOCHS on that thing!

    6. Re:the real question is... by Sycraft-fu · · Score: 2

      Probably not as many as you'd think. IT doesn't have any inherant 3d capabilities so you'd have to write a software mode OpenGL renderer. Ok, not a real problem as there is plenty of power to go around, HOWEVER the DSPs that we use to do graphics these days are really fast. YOu'd would probably need a computer in the multi-teraflop range just to emulate a GeForce 4. This is also assuming that graphics calculations like that can be doine in a massively parallel way, whichi I don't know if they can or not.

      A much more imperssive result would be obtained by talking to www.quantum3d.com and getting them to build you a system based on their ObsidianNV system.

    7. Re:the real question is... by Anonymous Coward · · Score: 0

      No, you're wrong. Compare the number of processors on the ES compared with, say, ASCI White, or ASCI Red. Then do some sums for performance per CPU.

    8. Re:the real question is... by Anonymous Coward · · Score: 0

      UNICODE works better

    9. Re:the real question is... by afidel · · Score: 1

      Probably not many. Computers of this size generally don't have exceptionally fast processors, just shitloads of them.

      Although maybe true for some systems it is most definitly NOT true for this one. Each of the 8 cpu's per node do 8Gflops, or about twice as fast as the fastest P4.

      --
      There are 4 boxes to use in the defense of liberty: soap, ballot, jury, ammo. Use in that order. Starting now.
    10. Re:the real question is... by vrmlknight · · Score: 1

      altho Quake 3 has the option to run to take advantage of SMP (altho i never got it to work with out crashing) but I think that this systems goes way beyond SMP

      --
      This must be Thursday, I never could get the hang of Thursdays.
    11. Re:the real question is... by notaspy · · Score: 1

      No. The real question is, does it have a two button mouse?

      --
      hi!
    12. Re:the real question is... by Black_Logic · · Score: 1

      I promised myself i'd stop commenting on .sigs but...

      Wouldn't it make more sense if you said, ("I don't get it.", Is not a rebuttal.(sp?) ) ?

      Also, spell rebuttal right, you'll get more credibility. :)

      --
      Ansi's and stupid tricks!
    13. Re:the real question is... by Lars+T. · · Score: 2

      Well, it's bad enough as a rebuttal, but some people actually try to start a discussion with: "I don't get how anybody would ...", without giving any reason.

      --

      Lars T.

      To the guy who modded me down from perfect to terrible Karma - Apple haters still suck

    14. Re:the real question is... by Black_Logic · · Score: 1

      ::Laugh:: I see, funny. Then, how about, --

      "Personal preference is not a good basis for an argument." :)

      Only to better clarify, you see; not because the original isn't insightful.

      (hope i'm not being irritating, I actually like .sigs. :) That couldn't be any worse than having an afinity for poetry, could it? :)

      --
      Ansi's and stupid tricks!
  18. Ahh.. by KarmaBitch · · Score: 0, Offtopic

    But, can it figure out why Cowboy Neal always wins the polls? Hmm... Ya know... Some things are probably best left unsolved.

  19. algorithm development by FrenZon · · Score: 3, Interesting

    This is a question, and not a statement

    While this does a nice job of crunching numbers, how do they know that their algorithms are any good at doing what they do? Or are they trying to simulate things that aren't continuously kicked around by chaos theory?

    I ask because I've been looking at dynamics in my spare time, and simulating something as small as cigarette smoke accurately seems impossible (although I must say Jos Stam and Co did a nice job of making it look real). So it seems a bit bewildering to see something trying to simulate the earth, even if only at a macro level.

    1. Re:algorithm development by Oily+Tuna · · Score: 2, Insightful

      simulating something as small as cigarette smoke

      Ah .... but they're simulating big things. Big things are easier to simulate than little things, not harder.

      --
      Mmmmmmm ... sushi.
    2. Re:algorithm development by mfago · · Score: 1

      They validate the algorithms using experimental data.

      The ASCI computers for instance are undoubtedly being validated against the last few nuclear tests.

      This is harder than it might sound.

      BTW "big things are easier to simulate than small things" (in another reply) is utter crap.

      Goes to show you though that using "off-the-shelf" hardware is not necessarily the best thing to do.

    3. Re:algorithm development by Anonymous Coward · · Score: 0

      I think what you mean is, more general things are easier to simulate than detailed things. Nothing to do with big or small.

    4. Re:algorithm development by lingqi · · Score: 1

      I also believe that they don't know if their algorithms are good, just as you said.

      one of the purposes they said was "Establishment of simulation technology with 1km resolution."

      Probabbly use some old data and see if they can predict a little bit into the future with reasonable accuracy.

      as a side note: on the "problem it solves" page, notice how many "seismicity" related items there are (% wise)! i think the true reason they want this thing is to predict earthquakes. i don't blame them; Tokyo is expecting a "big hit", and considering almost 1/2 of the japanese population lives there, probabbly even more than that in terms of $$ in japan -- it's a good idea they are trying to predict these things.

      p.s. Japan experiences small quakes almost daily -- most of them cannot be felt; but it means they have tons of data to verify their simulations against.

      --

      My life in the land of the rising sun.

    5. Re:algorithm development by lowLark · · Score: 5, Informative

      The easiest way to validate these types of prediction mechanisms is to feed them only part of your data set and see how well it predicts your remaining dataset. For example, if you have an ocean temperature data set from 1920 to the present, you might start by feeding it 1920-1992 and seeing how well its predictions for then past ten years hold up to you actual data. You may think that the known data set it too small for accurate predictions, but there are some fascinating methods (like ice core sampling and tree growth sampling) that seem to allow pretty good deductions as to past climate conditions over a very long period of time.

    6. Re:algorithm development by Anonymous Coward · · Score: 0

      seismicity...

      Yeah; that would also explain 'Earth' Simulator...
      For that they really need to simulate (the physics & mechanics of) Earth...

      Good luck!

    7. Re:algorithm development by Anonymous Coward · · Score: 0

      ...and considering almost 1/2 of the japanese population lives there...

      Who told you that half the Japanese population lives in Tokyo?!?

      Japan has around 120 million people in total - do you mean to say you really believe Tokyo has 60 million people? F'crying out loud, go back to high school and redo basic geography.

      (For those of you who'd like to know, Tokyo has around 12 million people during the night, and 20 million during the day - more if you consider the entire Tokyo-Yokohama metropolitan area as one city.)

    8. Re:algorithm development by hype7 · · Score: 3, Insightful

      While this does a nice job of crunching numbers, how do they know that their algorithms are any good at doing what they do? Or are they trying to simulate things that aren't continuously kicked around by chaos theory?


      This is an extremely insightful comment.

      What's being suggested here is akin to this - sure, they've got the most powerful car in the world, but to get from LA to New York, you've got to head east. Heading North won't help much, no matter how powerful your car is.

      This is what gets me about all these global warming "earth is going to heat up and cool down and rain and drought and..." predictions. How can they be sure they're even in the ballpark?

      One variable out, they could throw their predictions out by a massive amount. Their simplifications to allow for the computer to do predictions may not take into account the nuances and subtleties of the real world.

      That's why, in many instances, I look at these computers with perhaps more cynicism than most other people. They're great for testing theories, and for allowing scientists to compute algorithms that they possibly otherwise wouldn't be able to do. But just because it's come out of a billion $$$ computer, doesn't mean it's a golden egg.

      It's like that old saying that came out when word processors were first invented - shit in, shit out. Just because it's been through a fancy (or expensive) machine it doesn't make the outcome any more valid.

      -- james
    9. Re:algorithm development by Anonymous Coward · · Score: 2, Interesting

      Nowadays, when similating complex environments such as weather systems, the main innacuracy comes from now knowing the "starting state" with enough precision. Obtaining wind, temperature and pressure information is easy for the data points that lie conveniently on the surface of a landmass, but data points way up in the air or out to sea are mostly calculated through interpolating known points. I know that in 1992, the Cray YMP12/128 was doing 10-day predictions of weather with a 320x640x31 matrix of data points and a 15 minute time-step - there's no way it would have had accurate data for many of those points at all. The similation took 6-8 hours - roughly a third of that was pre-processing to compute starting conditions, another third for the time-stepping simulation, and the final third for post-processing to derive qualitative conditions.

      The accuracy of the simulation can be measured in terms of the length of time that the predictions remained within a given error of the actual weather.

      To overcome the problem of inaccurate starting states, high performance computing is now used to run many simulations of the same thing in parallel, each with a slightly different starting state. The hope is to identify many of the "exceptional" outcomes, and assign a probability to that outcome.

      A good example of this is the October 1987 storm in the UK, which the Met Office didn't see coming at all. It is believed that had they been able to run many simulations with different starting states, they would have seen that starting conditions slightly different from those used in their simulation would have lead to the craziness that ensued.

      More information about the storm and its cause can be found here or in the Google cache.

    10. Re:algorithm development by Anonymous Coward · · Score: 0

      Yes, this is absolutely true.

      They are trying to model climate change, not the weather. Everyone knows that you cannot predict the weather far in advance because of chaos - you need exponentially greater accuracy of the initial data to get linear improvement in the prediction.

      The assumption the climate modellers are using is that, although the _weather_ is chaotic, the _climate_ isn't. ie, somehow the chaotic perturbations affecting the weather get 'washed out' in the large-scale structure of the Earth's climate.

      There is quite a large body of academia that thinks that that assumption is wrong, and that the Earth Simulator will not produce any useful science. But there is very little justification either way.

    11. Re:algorithm development by TonyJohn · · Score: 1

      Nah. Big things are easier to simulate than little things. You can ignore quantum properties for one thing.

      The problem comes when you try to simulate LOTS of little things. The Earth falls into this category.

      --
      Owl tried to think of something wise to say, but couldn't.
    12. Re:algorithm development by Misanthropic+Lycanth · · Score: 1
      Ah .... but they're simulating big things. Big things are easier to simulate than little things, not harder.

      Not exactly true. Simulating large systems with the same level of accuracy as small systems is definitely harder. They will probably making a perturbative analysis of climate as opposed to asking where every single Nitrogen molecule will be.

      --

      Physics: Making the universe open source.
  20. Nothing New by RAMMS+EIN · · Score: 2, Funny

    ``The Earth Simulator Project will create a "virtual earth" on a supercomputer to show what the world will look like in the future by means of advanced numerical simulation technology.''
    We already have that: http://whatisthematrix.warnerbros.com/

    --
    Please correct me if I got my facts wrong.
    1. Re:Nothing New by Anonymous Coward · · Score: 0

      "the 13th floor" came before the matrix. A step over the Matrix it had was that it wasn't a simulated world which contained humans dynamically inserted into it. EVERYTHING was artificial, from the people, their personalities, through to the whole darned world.

  21. Mandatory remark.... by Dark+Lord+Seth · · Score: 0, Redundant

    Imagine a Beowulf cluster of these !

  22. Earthquakes shmirthquakes. by ErikTheRed · · Score: 5, Funny

    They don't want to admit it, but the real reason for building this thing is so that they can predict appearances of Godzilla....

    --

    Help save the critically endangered Blue Iguana
  23. Re:Why? by Anonymous Coward · · Score: 0

    Hahahahahaha

  24. "Advantages" of ES by binaryDigit · · Score: 5, Interesting

    Seems to me that though ES takes the overall performance crown, that the IBM and HP (man that sounds strange) units have some definite advantages over it. Primary of which is the fact that they DO use "off the shelf" parts. ASCI White uses 375Mhz Power3 chips which are comparitively low performance compared to what IBM is shipping now (1.3 Ghz Power4). I don't know what the technical details are behind ASCI White, but it seems that IBM could instantly get a doubling of performance by using new CPU modules. With the "specialized processor" approach that NEC uses, this would seem to be prohibitively expensive. IBM has already amortized most of the cost of the development of new processors through their normal business units.

    Another advantage would be that since ASCI White is a hyper RS6K, you could use a lower end model (and IBM could rather inexpensively offer a lower end model) to develop your models on before using the relatively expensive big boy to do the actual simulations. I have to admit that this point is moot if they don't keep the utilization of the thing up pretty high most of the time.

    Though they mention that ES "only needs 5104" processors vs 8192 for AW, it looks like ES still takes up massive amounts of space. Now ES' storage is significantly larger that AW, so maybe that's where all the space is being eaten, but it would be interesting to see what the actual cabinet space/power requirements for the two machines sans storage are (assuming they are both using standard stuff for storage).

    Others things include since AW is based on OTS parts, is it easier to get parts for when processing units konk out. Is it simpler for a tech to work on the unit. Since Linux is already running on RS6K, theoretically with a few device drivers, you could run Linux on that bad boy

    Of course all this is moot in the non-real-world of supercomputers. With seemingly infinite budgets, the only _real_ measure is absolute performance, and ES obviously has the edge here. But if I were the IBM sales rep for supercomputing, I'd sure be hyping the fact that when it's not simulating nuclear explosions, you can run Gimp and Mozilla.

    1. Re:"Advantages" of ES by lingqi · · Score: 1

      ...compared to what IBM is shipping now (1.3 Ghz Power4)

      Wow... -- somebody else who knows about the power4...

      I am cuious, do you know how they USE the darn things? i mean, the sucker has over 5,000 pins (!!), i suppose the thermal requirements are tremendous too. any info would be appriciated.

      But if I were the IBM sales rep for supercomputing, I'd sure be hyping the fact that when it's not simulating nuclear explosions, you can run Gimp and Mozilla.

      don't forget to mention the terrific pr0n potential.

      --

      My life in the land of the rising sun.

    2. Re:"Advantages" of ES by alfadir · · Score: 3, Informative

      You can get an SX-6i . The processor in ES is not made only for ES. And I don't think you would sell many supercomputers for IBM if you were advocating Gimp and Mozilla as applications...

    3. Re:"Advantages" of ES by binaryDigit · · Score: 2

      You can get an SX-6i

      True, but if you look at the Top1000 list, you'll see significantly more IBM machines across the board then NEC, including a large number of "standard" units (sold as kick ass RS/6000's vs "supercomputers", e.g. the P690

      I would think that this gives them a signficant edge in development costs as well as giving their customers more flexibility.

      And I don't think you would sell many supercomputers for IBM if you were advocating Gimp and Mozilla as applications

      Oh come on, nuclear physicists like to clean up photos of their dogs (probably don't have girlfriends) and surf the web just like anyone else ;) Imagine the speed in which those nerdy scientists can apply those Gimp filters to all that pr0n they download.

    4. Re:"Advantages" of ES by peatbakke · · Score: 4, Informative

      I'm not sure how much you've looked up, so some of this information may be redundant, but here's what I've been able to dig up:

      • The Power4 chip has two processor cores and a shared 1.44MB on-chip L2 cache .. which in turn appears to be implemented as three separate cache controllers. The cache lines are hashed across the controllers, which is a pretty neat trick IMHO.
      • It weighs in at about 170 million transistors
      • This PDF mentions that there are over 5500 total I/Os (including > 2200 signal I/Os) that give the chip a raw bandwidth of over 1 Tb/s.
      • According to this page, the chip simulations show the core temperature peaking around 82C (~180F) in certain regions of the chip, and consuming 115 - 140 watts.

      That's a beast of a chip! The packaging looks pretty substantial as well. I don't doubt the cooling systems are fairly remarkable, although I can't find any specific information about 'em.

      cheers!

    5. Re:"Advantages" of ES by Abnormal+Coward · · Score: 1

      Does it use a floppy disk drive for its L2 storage ?

    6. Re:"Advantages" of ES by nr · · Score: 2, Informative

      The ES is running a standard UNIX called Super UX which I guess is fully POSIX compliant and have the normal ANSI/ISO C compiler. You should be able to compile and run most Open Source programs including Gimp and Mozilla. I have compiled and used many GNU and Open Source programs like Emacs on the universitys Cray running UNICOS which also is a UNIX derviate designed for the Cray vector supers.

    7. Re:"Advantages" of ES by fb · · Score: 2, Insightful

      >you'll see significantly more IBM machines
      > across the board then NEC

      As widely known and publicly acknowledged also by one of the authors of that list (Prof. Meuer), the linpack benchmark used to build the top500 list is biased against vector supercomputers, like NEC's.

      Supercomputer performance cannot be measured by a single number, really.

      --
      fB
    8. Re:"Advantages" of ES by fb · · Score: 1

      Super UX is an almost stock version of System V.

      --
      fB
    9. Re:"Advantages" of ES by wsloand · · Score: 1

      But if I were the IBM sales rep for supercomputing, I'd sure be hyping the fact that when it's not simulating nuclear explosions, you can run Gimp and Mozilla.

      Supercomputers aren't for anything but processing. Why would someone care to run something like gimp or mozilla on a supercomputer, or anything at all other than simulations? The purpose of these massive-scale systems is strictly computation.

      If someone is regularly using a supercomputer for Gimp or Mozilla, then there is something wrong.

    10. Re:"Advantages" of ES by binaryDigit · · Score: 2

      But if I were the IBM sales rep for supercomputing, I'd sure be hyping the fact that when it's not simulating nuclear explosions, you can run Gimp and Mozilla

      People, people, this was a joke. You know, not intended to be taken seriously. Of course if someone is going to spend 10 figures on a computer, they don't give a flip about Gimp, etc. Chill, it's ok, put the Pepsi down.

    11. Re:"Advantages" of ES by Traksius+Egas · · Score: 1

      Speaking of being biased or having an edge, I'm curious if the top500 takes into account processors with multiple cores in one package? I believe the Power4 has two cores. Would that be one processor or two?
      ASCI-W uses the Power3 (I believe) so switching to the Power4 would theoretically double the performance without increasing the processor count.
      Does the SX-6i fit into that category also or are vector procs not really built the same way?

    12. Re:"Advantages" of ES by lingqi · · Score: 1

      heh... thanks for the link

      actually the Power4 is, i think, even more complicated than that: the chip is not put into a system alone, they are stuck together onto a multi chip module that has 4 chips (while each chip has 2 processor cores.) so each MCM is already 8-way SMP'ed. and *then* they have to cool this thing, 4 massively power-hungry chips in such close proximity? wouldn't want to think about it.

      and oh yeah... 32M L3 cache... i remeber back when my PC had only 8M of MAIN MEMORY...

      p.s. on the PDF, there are some pages that said "IBM confidential." haha.

      --

      My life in the land of the rising sun.

    13. Re:"Advantages" of ES by some2 · · Score: 0

      From the spscicomp.org PDF, on the 14th page, it was noted that this chip could boot with Linux, AIX, and OS/400 and worked with SMP. Is this simply due to IBM's willingness to work with Linux, or do other mainframe CPUs work with Linux using SMP?

      On the side --- I'd love to see the kernel compile on that beast!

    14. Re:"Advantages" of ES by peatbakke · · Score: 1

      Yeah, the MCM is a real beast -- It's about the size of a grilled cheese sandwich, which seems like a more-than-convenient coincidence. ;)

  25. I'm in luck. by Oily+Tuna · · Score: 0, Troll

    I know the cheat codes!

    Hmmm blonde or redhead first?

    --
    Mmmmmmm ... sushi.
  26. Parallel computing != distributed computing by Anonymous Coward · · Score: 0

    That "shitload of CPU power" comes with an expensive highspeed network which you don't need for distributed computing applications.

  27. whatever the case is... by systmc · · Score: 1

    Seems kinda silly if we already know what the answer is. Hmm... double checking, perhaps?

    1. Re:whatever the case is... by Sheridan · · Score: 2
      Seems kinda silly if we already know what the answer is. Hmm... double checking, perhaps?

      Ahh, but you're forgetting that the Earth wasn't designed to calculate the answer, (Deep Thought had, as you rightly note already told us what that is) it was designed to calculate the question.

      Would've worked, too if the pesky Golgafrinchans hadn't turned up and perturbed the calculations. By the time the Vogons demolished it, the algorithms were way out of whack anyway.

      (Yeah, I know: -1, Offtopic)

  28. Can you imagine... by bakes · · Score: 5, Funny

    ..a single-cpu one of these!!!

    --
    Ho! Haha! Guard! Turn! Parry! Dodge! Spin! Ha! Thrust!
    1. Re:Can you imagine... by Anonymous Coward · · Score: 0

      You would have to plant that darn thing on the southpole in order to cool it. But then you couldn't run it for very long ... since the ice is melting one way or the other.

      cowardly resonding,
      yours,
      the dude

  29. Ahh, it may be all powerful... by AussieBastard · · Score: 1

    ...but will it run pong? :)

  30. Ah, but does it use... by Anonymous Coward · · Score: 0

    .. the CrystalSpace 3D engine?
    I thought not, it isn't powerful enough.

  31. For the benefit of humanity! by opti6600 · · Score: 0

    I personally agree with you one hundred percent in regard to your thoughts. All of these supercomputers dedicated to abstract tasks like climate forecasting, protein folding, and things of that nature are all and good, but they do very little to help the human race as it is, right now. I propose that you take Blue Gene, and the Earth Simulator, and ASCI White (which is used for war of all things), and port clients for distributed computing applications to them. My proposal is based primarily on the United Devices Cure for Cancer project. Take this computer, let it run for a few days, and you have yourself a perfect list. The same procedure could be applied to AIDS, anything, as long as someone could write a similar program for it. It's great that Blue Gene is helping scientists learn against their will about protein folding, and that information could very well help us in the future. The problem however lies in the fact that this is a truly massive amount of processing power that could be put towards better use. You could research anything, just get three teams: -Team One: researches applications for the processing power of a computer that scale, that pertain to the immediate survival and improvement of the human race -Team Two: builds a distributed computing application to solve the problem that Team One comes up with -Team Three: applies the regular x86 compilation of the Team Two distributed computing applications to supercomputers For instance, you could use Blue Gene to solve work units, and something like ASCI White to feed them to it. But seriously people, we need to rig something that helps us, not just wastes taxpayer or private sector money on another extraneous project.

    1. Re:For the benefit of humanity! by Anonymous Coward · · Score: 0

      I agree, but I'm sure they've already taken that into consideration and found it to not be in _their_ best interests. If we _REALLY_ wanted to find a cure for cancer, find aliens, or whatever, people would pool their resources and develop several of these supercomputers and get them cracking on the problem. This is a possibility given the amount of wealth floating around these days.

      But do we see this happening? Nope. Why? Because humans are greedy little shits that are only concerned about something once it effects them in some way.

  32. by the way by lingqi · · Score: 1

    ES has 700TB storage, ASCI W 160TB;
    a small difference... heh

    --

    My life in the land of the rising sun.

    1. Re:by the way by binaryDigit · · Score: 2

      Yes, I mentioned the difference in my post:

      Now ES' storage is significantly larger that AW ...

  33. I clicked on the link by hype7 · · Score: 1

    but I couldn't find anything about the PowerMac G4?

    -- james
    ps humour, not troll/flamebait :)

  34. Good Ol' Days by Ami+Ganguli · · Score: 2

    Back in the BBS days you could enter [Control-H]'s into a message and they would become part of the actual text. That way the reader could actually see the words appear and be back-spaced over and re-written. It was a cool effect. Of course it worked better at 300-2400 baud, where you could actually see the characters being drawn.

    --
    It is tempting, if the only tool you have is a hammer, to treat everything as if it were a nail. - Abraham Maslow
  35. We already know the answer by saphena · · Score: 0, Redundant

    42

    1. Re:We already know the answer by Anonymous Coward · · Score: 0

      The earth was not build to give us the answer but the question itself.

  36. "I predict by zephc · · Score: 3, Funny

    that within 100 years computers will be twice as powerful, 10,000 times larger, and so expensive that only the five richest kings of Europe will own them." - Prof. Frink

    --
    "I would say that 99 per cent of what my father has written about his own life is false." - L. Ron Hubbard Jr.
    1. Re:"I predict by Anonymous Coward · · Score: 0

      Ha ha

  37. That's nothing! Check out the Playstation 3! by Pilferer · · Score: 1

    Where have I seen this before?

    http://www.geocities.com/s178.rm/index2.html

  38. Unfortunately... by techmuse · · Score: 2

    The Earth simulator will be destroyed to make way for a hyperspacial bypass...

    1. Re:Unfortunately... by Ralph+Bearpark · · Score: 3, Funny
      The Earth simulator will be destroyed to make way for a hyperspacial bypass...

      Well, it would be kinda ironic if it got knocked out by an earthquake. Especially if it didn't predict it.

      Regards, Ralph.

    2. Re:Unfortunately... by The+Creator · · Score: 1

      >Well, it would be kinda ironic if it got knocked out by an earthquake. Especially if it didn't predict it.

      It whould be even funnier if it did predict it. And committed suicide.

      --

      FRA: STFU GTFO
  39. Re:You AC's have no hope of winning. by Anonymous Coward · · Score: 0

    Get yer tits out and I'll think about it.

  40. Do they have the way out? by littleRedFriend · · Score: 0, Redundant

    Come on give it try and install windows on it. Ooops it went just down from 40 teraflops to effectively 1 (the other 39 is probably used for IE).

    --
    IANAL, but imagine a beowulf cluster of in Soviet Russia all your belong are base to us welcoming the new SCO overlords.
    1. Re: Do they have the way out? by Anonymous Coward · · Score: 0

      Shut up.

      :)

    2. Re:Do they have the way out? by netsharc · · Score: 1

      Well, they probably could, but at least we have a world-wide group of geeks that did just that and afaik they're not all smelly, dirty or losers

      --
      What time is it/will be over there? Check with my iPhone app!
  41. This would be a good machine to... by Anonymous Coward · · Score: 0

    Overclock! Why not? It's probably got an excellent cooling system. Someone needs to write up a guide on [H]ardOCP or something...

  42. alright now. by domtropen · · Score: 2, Insightful

    i'll freely admit that i'm a little freaked out by this. (so much so that i'm delurking.) directed at the discussion about whether it can be a practical machine, does it really matter? it seems as if it was built for one sole purpose, and it appears that it will do it well. can we just give them that?

  43. SETI? by Anonymous+Squonk · · Score: 2, Interesting

    Are there any estimates of the processing power of all the worldwide computers participating in the SETI project?

  44. Topic by sheepab · · Score: 2, Funny

    Inside The World's Most Advanced Computer

    How did they get inside my brain???

  45. Earth Simulator OS + German TV by alfadir · · Score: 5, Informative

    The Earth Simulator is running Super UX. The same operating system as the rest of the NEC supercomputers

    The German Language TV channel 3sat will broadcast a 30 min film on Earth Simulator on Monday and 24th of June at 21:30 hours and on Tuesday, 25th of June at 14:30 hours.

    1. Re:Earth Simulator OS + German TV by Anonymous Coward · · Score: 0

      Nice, so could you record the program and put it up on some file sharing app for the rest of the world to see?

  46. Uhh ... by Anonymous Coward · · Score: 0

    Dammit. I thought one day my little university would also be able to get into the top500 list since it is bulding an Sun Fire Ultra Sparc Cluster (http://www.rz.rwth-aachen.de/hpc/sun/index_e.html ). The setup will consist of 16 Well know the world knows :)

  47. HAHAHA by autopr0n · · Score: 0, Offtopic

    It's funny because windows is slow!

    --
    autopr0n is like, down and stuff.
    1. Re:HAHAHA by Anonymous Coward · · Score: 0

      No asshole, it's funny because Billy Boy claims he can do this.

  48. NSA by SJ · · Score: 1

    ES seems all very interesting and all, but I would like to see how it compares to what the NSA has parked under Washington for Echelon and its successors.

    You can get your butt that if information about the "Worlds Fastest Supercomputer" is available to the general public, the NSA has got something bigger and better.

    1. Re:NSA by Arcturax · · Score: 2

      I don't mean to nitpick, but the general public doens't get access to this. Researchers will, but you can't just walk up and start playing Doom on this thing.

      --

      --Won't that be grand? Computers and the programs will start thinking and the people will stop. - Dr. Walter Gibbs
    2. Re:NSA by SJ · · Score: 1

      I know we don't get to play with it, but simply the fact that we _KNOW_ about it, means that the various security organizations out there have something bigger and better.

      It's like the Stealth Bomber. The public only found out about it because it was obsoleted by a couple of guys in Perth, Western Australia who invented a type of radar that would pick it up.

      As a general rule, the public only find out about this stuff when there is something better.

  49. Uh ... no german university in the top500 ... by Anonymous Coward · · Score: 2, Interesting
    Oh well, looks like my university is never going to make it into the top500 list of super computers (not to speak of any other german university).
    Although they are setting up a quite cool Sun Fire Ultra Sparc Cluster running Solaris.

    The setup will consist of 16 Sun Fire 6800 SMP nodes (1500 MHz, each node is a 24 processor SMP system with 24 GB shared main memory) and 4 Sun Fire 15K SMP nodes (1500 MHz, each having 72 processors and 144 GB of memain memory) giving an max. arithmetic performance of 4 TFlop/s.
    Check the link to see for yourself (like you dont have anything better to do, right?).

    Sad/funny part of the story: the cluster is going to be finished in 2003 ...
    I should check Moores law on top 500 super computers...

    Alt least know the world knows we do cool stuff too ...

    1. Re:Uh ... no german university in the top500 ... by Anonymous Coward · · Score: 0

      It'll never make the TOP500 because the Solaris Operating system doesn't scale well.

  50. Yes but does it run Microsoft Word ?? by derekb · · Score: 0, Redundant

    Sorry had to ask..

    Seriously though, the 40 TFLOPS is a peak value. On well-written-code, I wonder how fast it's capable of versus the number 2 in a similar comparison.

    Saying 40 TFLOPS peak is still too much like 'Hey my 2.2 GHz Intel beats your 1.2 GHz AMD'. While it is fast, is it really that accurate?

    1. Re:Yes but does it run Microsoft Word ?? by fb · · Score: 1

      A test of a real application showed a performance of 28 Tflops sustained (70% efficiency). Not at all bad.

      --
      fB
    2. Re:Yes but does it run Microsoft Word ?? by kkkalen · · Score: 1

      And if it does, will you still see the "File too large for Notepad" message on a regular basis?

      This is a rhetorical question?

      --
      If you don't believe me, ask that guy over there.
  51. It runs mainly FORTRAN? by MaSee · · Score: 1

    Almost 'nuff said. But I guess for the number crunching work that it's designed to do, FORTRAN must be the way to go. Not to mention that it was probably build by engineers. Nya! ^_^

  52. Re:What about by Anonymous Coward · · Score: 0

    The cuntface reply wasn't exactly a surprise either.

  53. Re:Wow... by Anonymous Coward · · Score: 0

    A possibility, yes, but in reality there's only me. And I don't really read ... I sort of skim and then reply regardless.

  54. Food for thought... by gentix · · Score: 1

    That's probably true, since the sum of all processing power in silicon valley of about 25 years ago is now wrapped up in a single personal computer.
    weird huh?!

  55. Re:Why? by Anonymous Coward · · Score: 0

    boo hoo the americans got smashed at their own game. stupid clown. obviously you're much too smart to be able to build one of these. yep, anything about 10TFlops is just stupid because you can just dig a hole.

  56. Re:What about by Anonymous Coward · · Score: 0

    And they say /. is boring.

  57. SimEarth, SimCity and Sims by Bubblesculpter · · Score: 1



    Wonder if EarthSimulator could run a version of SimEarth, down to every country,state,city,person,etc. Doesn't the Sim series also throw in weather events? Tornados, etc.? EarthSimulator should be able to crank out a few of those...

    --
    www.Beyond7.com Insane modern art water sculpture.
  58. /.'ed by cibrPLUR · · Score: 2, Funny

    the new list of supercomputer rankings is up today.

    I guess top500.org isn't running on one of them.

    --

    -cibrPLUR

  59. My old computer was more advanced. by Anonymous Coward · · Score: 1, Funny

    I spilled a cup of coffee into it.

    It didn't fry.

    Beat that, Earth Simulator! Beat that!

    (I still use the now-slightly sticky soundcard from it :P)

  60. So... by StoatBringer · · Score: 0, Redundant

    Can they play Quake III on it at lunchtime?

    --
    Cress, cress, lovely lovely cress
  61. Divine intervention? by kirkunit · · Score: 1

    I wonder just how well it can simulate Acts of God(tm). Or maybe they can tap into the Global Consciousness Project?

    1. Re:Divine intervention? by The+Creator · · Score: 1

      Hey, as far as slashdotters are concerned that thing IS God.

      --

      FRA: STFU GTFO
  62. stupid reference right below this.... by DanThe1Man · · Score: 3, Funny

    The Earth Simulator Project will create a "virtual earth" on a supercomputer...

    Hmmm, now where have I heard of an idea like that?

  63. "Inside" the Earth Simulator by erlando · · Score: 4, Interesting

    There are some nice pictures on the ES site as well. I wonder if the colouration of the cabinets is there to prevent the engineers from getting lost..? :o)

    --
    Remember, there are no stupid questions. But there are a lot of inquisitive idiots.
    1. Re:"Inside" the Earth Simulator by hyoo · · Score: 2

      A better view of the layout is shown on the bottom right of this page, the purpose of the color coding appears to be for aesthetics. You know, for when godzilla rips the roof off the building.

    2. Re:"Inside" the Earth Simulator by caveat · · Score: 0, Troll

      i really like the design aesthetic of the cabinets, with the big red circles at the ends...remind anybody else of HAL, or pretty much every menacing intelligent supercomputer ever conceived? makes the ASCI systems look so boring, with plain metal cabinets...gotta love the japanese Way, everything has to look good on top of working good.
      now if they'd just backlight the circles deep red, and pulse the light along with the level of system activity...mmmm

      --

      Facts do not cease to exist because they are ignored. - Aldous Huxley
    3. Re:"Inside" the Earth Simulator by Mr.Sharpy · · Score: 1

      It's quite lovely. However with such attention to detail in the system, they should have spent a little more on room it is housed in. Just look at that lighting scheme! Where's the trendy halogen spot downlights? Where's the accent lighting around the round thing in middle of the room? My god man, how can they expect engineers to slave in such conditions! Oh the humanity! Flourescent row lighting! And you just know they make the buzzing sound...

    4. Re:"Inside" the Earth Simulator by Quazi · · Score: 1

      Imagine this as a 3U rackmount case.. heh!

    5. Re:"Inside" the Earth Simulator by Buck2 · · Score: 1

      It actually reminds me of the Japanese flag, not HAL. But now that you mention it, no, it doesn't remind me much of HAL, it still makes me think of the Japenese flag.

      --

      As my father lik@(munch munch)... ....
  64. What about the REAL fastest computers? by Alsee · · Score: 2

    I was looking at the list of supercomputer rankings, and I couldn't help thinking - yeah, but what about all the CLASSIFIED computers? I bet the US gov has secret computers that would blow that list away.

    -

    --
    - - You can't take something off the Internet! That's like trying to take pee out of a swimming pool.
    1. Re:What about the REAL fastest computers? by fb · · Score: 1

      They are there as well. The classified computer project of the US government is named ASCI...

      --
      fB
    2. Re:What about the REAL fastest computers? by caveat · · Score: 0, Troll

      They are there as well. The classified computer project of the US government is named ASCI..

      well...ASCI is publicly acknowledged computers running classified apps/data. i wouldn't say it's totally unreasonable to say that the NSA/CIA/MI6/etc likely have some non-publicly-acknowledged system(s) hidden away somewhere that make ES look like a 8088. i don't have that much faith in my government to expect them to tell me everything they spend my tax dollars on.

      --

      Facts do not cease to exist because they are ignored. - Aldous Huxley
  65. SUX by hey · · Score: 1

    So they are running SUX - has it occured to them that it sounds like SUCKS? Maybe they should take a lead from the former the canadian "Conservative-Reform Alliance Party" ... yes C.R.A.P. and change their name

    1. Re:SUX by foobar104 · · Score: 2

      Naming their OS S-UX would be pretty much par for the course for a Japanese company. A few years ago, Sony took some of their videotape library technology and applied to to data storage. The video version was called "TeleFile," I believe. They decided, since the library could hold as much as a petabyte, to call the data version "PetaFile."

      Shortly thereafter, Sony started referring to the libraries as "PetaSite" systems instead. Say "PetaFile" out loud, and you'll understand why.

      I'd provide a link, but Sony's web site works properly in, like, no known browsers. Pfeh.

    2. Re:SUX by modecx · · Score: 1

      Similarly, Toshiba was working on a web spider/search engine back when the www was becoming more public. Apparently, the Japanese people love Woody the Wood Pecker, and Toshiba adopted Woody as their mascot for this site, and referred to Woody as "Woody the Internet Pecker". Eventually, word of the English meaning of this got around to one of the bigwigs at Toshiba, and the whole project was cancled.

      Amusingly, there are many stories like this floating around somewhere. I've forgotten most of em, sorry to say.

      --
      Constitutional rights may be respected, repealed, or modified; but they must never be ignored.
  66. Wow by Ilgaz · · Score: 2, Funny

    What I read is, its based on Nec SX arch, now imagine if it was a DX...

    Some nostalgia ;)

  67. wow... by John+Pfeiffer · · Score: 1

    Boy would I like to buy some render time on THAT O_O

    (The article has some nice graphics too!)

    And because I KNOW there's probably 2000 "A beowulf cluster of those..." posts that are below my threshold: I would fear for the safety of mankind if someone made a cluster of /those/.

    --

    Friend: "The NIC is misconfigured..." Me: "No prob, I'll just telnet in and fix it." *Silence*
  68. Looks like a big PC right now... by Morky · · Score: 1

    The diagram that the post links to looks oddly like a big PC case. Even the way the internal components are layed out is oddly reminiscent of a standard PC.

    1. Re:Looks like a big PC right now... by cetan · · Score: 1

      Except they move the RAM in with fork-lifts :)

      --
      In Soviet Russia...michael would be rotting in Siberia!
  69. Wait a minute... by Zenex13 · · Score: 1

    from the preview-of-2026-ipaq dept.

    How can that be possible? You mean PocketPC OS will actually become powerful enough for that? You wouldn't be able to boot it up until AFTER you hacked it to run *nix.

    On the other hand, that would make a nifty graphing calculator. Wish i has a cluster of... ok i'll shut up.

  70. Re:algorithm development : real world results. by fw3 · · Score: 2
    While this does a nice job of crunching numbers, how do they know that their algorithms are any good at doing what they do? Or are they trying to simulate things that aren't continuously kicked around by chaos theory?

    Just because it's been through a fancy (or expensive) machine it doesn't make the outcome any more valid.

    Modelling real processes is a science which has been around for as long as computation. Simulations I used to run with Dynamo (discrete simulation of general PDE's) on a minicomputer was in some ways the coolest. It was also the slowest, a 10-state thermal transfer model could take an hour on a $200k processor.

    It is quite possible to look at fine-grained results using finite element or finite-difference methods in mechanical and fluid dynamics problems. For instance looking at vortex-shedding is within the realm of possible for a current model PC or workstation.

    verification is done against known data-sets and most simulation work involves checks on accuracy.

    Yes, problems which are really in the 'butterfly effect' region are very difficult, interesting (useful) work has been done taking such phenomena to the molecular level. For something like crack-propagation finite element methods have to be very detailed indeed to be predictive and while you can use these for useful results, the 'interesting' part needs to be calculated at the atomic level. That, however I have only seen done in simulation of highly regular materials.

    Many of the chaotic results happen where there is a delicate ballance in total energy, e.g. the dynamics of cigarette smoke. 'Useful' problems however usually involve substantial energy transfers and at some computational scale these are not chaotic.

    Solar and geo-thermal energy input into global weather patterns involves a LOT of energy and modelling is generally easier where you are looking at such problems.

    Computational weather prediction has made impressive strides. 10 years ago the ability to predict weather in New England was dismal, today between better sensors and better models the 5-day forcast is now more often correct than not.

    --
    Linux is Linux, if One need clarify their dist: <Dist>/GNU Linux
    bsds are of course just BSD
  71. I dont think my university ... by Anonymous Coward · · Score: 0
    I dont think my university RWTH-Aachen (not to mention other German universities) is ever going to make it into the list of top 500 super computers.

    Altough the setup is quite cool I guess you've never heard of it before.
    It is a Sun Fire Ultra Spac Cluster.
    Consisting of 16 Sun Fire 6800 SMP nodes (1500 MHz, each containing 24 processors, with 24 GB shared main memory per node) and 4 Sun Fire 15K SMP nodes (1500 MHz, 72 processors and 144 GB of memain memory each) it's arithmetic performace is to be about 4GFlop/s. Total memory 960GByte

    Too bad the whole thing only going to be assembled completly in spring of 2003.

    I must check wether Moore had a law on the entry level of GFlop/s for that darn list or not.
    Maybe we still get a chance.

    At least now the world knows that we do cool things too ;-o

  72. I quit reading here: by Zerelli · · Score: 0

    " Before this the aggregate performance of all 500 system was smaller THEN 30TF/s." I find it difficult to give the author any credibility when he cannot tell the difference between then and than.

  73. supercomputer = greatert than teraflop by peter303 · · Score: 2

    Now there are 23 installations at least this large, that should be the new threshhold: the largest machines that can be bought at any given time.

  74. The real advance is... by CHUD-Wretch · · Score: 0, Offtopic

    the big red eyeballs on the cabinets.
    As long as we can keep making such strides in Red Eyeball technology, HAL can't be very far off.

    No, really...there is a new lead dog...
    The amazing "Q" (built by CompaQ, go figure)

    ASCI Q, named to follow LANL's tradition of alphabeticalnames for computers, will have 11,968 processors, 12terabytes of memory and 600 terabytes of disk storage. It willbe about two and one-half times as powerful as today's mostpowerful supercomputer, ASCI White at Lawrence LivermoreNational Laboratory in Livermore, California. It wouldrequire approximately 20,000 of today's state-of-the-art PCs,which are capable of about 1.5 gigaOPS, working veryclosely together to match the peak performance of ASCI Q.

    or read more here/a?

    --
    "Suburbia is where the developer bulldozes out the trees, then names the streets after them."
  75. that's nothing... by RatFink100 · · Score: 1

    Zaphod's briefcase had an entire simulated galaxy in it.

  76. Gotta be asked... by SillySlashdotName · · Score: 1

    Yeah it looks nice, and probably plays a mean game of Quake (pun intended!), but what is its Bogomips number? Riddle me that, batman!

    --
    Acts of massive stupidity are almost never covered by warranty. --me.
  77. About half a human brain worh of processing power by gregor-e · · Score: 2, Interesting

    Hans Moravec estimates it would take about 100 Trillion instructions per second to emulate the human brain. At 38 Tflops, Earth Simulator is in the ballpark. Maybe they should have called it human simulator, or just "Sim".

  78. well by Anonymous Coward · · Score: 0
    how would you organize a 5,120 processor system capable of 40Tflops, and of course don't forget about the 10TB of shared memory

    4TB for pr0n, 3TB for mp3s, and 3TB for warez.

  79. Their web server sux by inputsprocket · · Score: 1

    they should use it to run their webserver. It sucks like a banshee. Wonder what its uptime is?

  80. Re:SETI? nah, no way. by caveat · · Score: 0, Troll

    haven't we been over this ad nauseum? yes, the total power of all the SETI computers is much much greater than ES, but that's all individual machines grinding away on their little chunks of data that they periodically download. for massive data-intensive numbercrunching like ES was designed for, the interconnects become as important as the processing speed. try running a whole-earth climate sim over the SETI network, where you have to constantly feed each processor a stream of data, and watch your performance drop by a few orders of magnitude. (disclaimer: IANA supercomputer architect, so everything i've said could be wrong, but i think i get the meat of the matter.)

    --

    Facts do not cease to exist because they are ignored. - Aldous Huxley
  81. Re:About half a human brain worh of processing pow by caveat · · Score: 1

    yeah, so what?
    it's circuts are permanently laid out, burned into silicon. that's what IMHO makes us intelligent...not the raw processing power, but the abitily to dynamically reconfigure our networks on-the-fly. now, if somebody were able to build a few 100Tflop systems with neural networks and the ability to fuse the networks, decide what features to keep and lose, and repeat, THEN we're in trouble (sort of like the way the escaping robot 'learns', only with >100,000x the processing power.)

    --

    Facts do not cease to exist because they are ignored. - Aldous Huxley
  82. What they are really trying to do... by FearUncertaintyDoubt · · Score: 3, Funny

    ...is build a computer capable of withstanding a full slashdotting.

  83. beowolf by Anonymous Coward · · Score: 0

    can you imagine a beowolf clust...umm...nevermind

  84. Imagine... by Anonymous Coward · · Score: 0

    a beowolf cluster of Earth Simulators.

    mmmmmmmmm...... beowolf

  85. I can't believe you guys missed it! by Anonymous Coward · · Score: 0

    640 processor nodes, each consisting of eight vector processors are connected as a high speed interconnection network.

    640 processor nodes.... Ought to be enough for anyone!

    Badoum boum! Thank you, thank you, I got a million of em!

  86. Re:huh? by Anonymous Coward · · Score: 0

    SHHHHHH!
    You don't want grammar nazi to hear, do you?

  87. Probably much greater by pclminion · · Score: 2
    The setiathome website claims there are three million users. Guessing that each user has about 10 MFlops, this is a total of 30 TFlops -- about the same as the Earth Simulator.

    However the SETI network could never do what the ES does because although it is compute-distributed, the data is centralized, so the actual compute rate is limited by communication speed. And over the Internet that's really slow compared to real interconnect architectures for these sorts of applications. At least, until the Internet can compete with a multi-gigabyte-per-second local interconnect. Of course by then, the processors will still be outstripping the network, so you probably still wouldn't be able to do it.

  88. I bet... by lord_codah · · Score: 1

    We'll be wearing that as our watch in 10 years. PLUS!! I just bought a geforce4 ti4200...I think I can hang with the Earth Simulator.

  89. Seti@home vs Earth Simulator by KE1LR · · Score: 1


    It's interesting to note that SETI@Home maintained a >39 TeraFLOP rate over the last 24 hours according to their web page. The earth simulator is aiming for 40.

    I know this isn't strictly an apples-to-apples comparison by a long shot but it's kind of fun to compare the two numbers.

    -- Ken

  90. Are the Japanese ahead????? by exhilaration · · Score: 1
    Am I the only one that was a little freaked out when they announced ES - a computer significantly faster than any American supercomputer? Why hasn't this been brought up??

    Japan already dominates the electronic and automobile industries, and all of a sudden it has the most powerful computer on the planet! The United States (or American companies) lead the world in the production and sales of computers: IBM, HP, Compaq, Apple - these are the brands that are on desks across the world. When people in other countries think "computers", they think America.

    http://www.es.jamstec.go.jp/esc/eng/images/g7_7_4. gif

    Take a look at the above graph - Japan is finally ahead in a race that's been going on for decades.

    Come on, isn't anyone else a *little* disturbed?? We're no longer the best in an industry that frankly we've bet our future on, and by such a wide margin!

    This is NOT A TROLL, and I have nothing against the Japanese - they make great stuff, but the computer industry belongs to AMERICA, and all of a sudden it looks like we've got a SERIOUS competitor.

    1. Re:Are the Japanese ahead????? by Anonymous Coward · · Score: 0

      pffft - you're really a patriotic troll. Watch a little more Star Trek to understand the philosophie behind computers.

  91. Maxis Introduces......SimEarth XP!!!! by shoemakc · · Score: 2, Funny

    SimEarth XP
    System Requirements:

    40Tflop 5120 Processor Cluster

    10TB of System Memory

    256 Color Display

    4X Cdrom Drive

    Arctic Rated Parka

    "Sales thus far have been slow..." confessed Wright, "...however we're expecting at least one large customer in the coming months."

    -Chris

    --
    --an unbreakable toy is useful for breaking other toys--
  92. And the answer is.... by TibbonZero · · Score: 0, Redundant

    42!!!!

    --
    Tibbon
    tibbon.com
  93. get your VDR ready ;-) by mrq1 · · Score: 0

    an here are the timesets for this two
    events of the show

    1:15:24:2130:2200:30:5:Super UX - THE EARHT:
    1:15:25:1430:1500:55:5:Super UX - THE EARHT:

    if it does not appear on kazaa .. call me ;-)

    btw: http://www.cadsoft.de/people/kls/vdr/
    rulez *sfg*

  94. A rather large problem... by Anonymous Coward · · Score: 0

    ... is that this beastie will produce 100 petabytes of data a day. When announced at a meeting of supercomputer people in London last year, this produced a very, very nervous laugh. Shipping output to people who can make use of it is a problem. Then, archiving...

  95. DOD bmf computers by trybywrench · · Score: 1

    i wonder if the DOD will want to use their nuke simulators on the earth simulator to simulate a nuclear holocaust... my guess is they probably don't want to know.

    --
    I came to the datacenter drunk with a fake ID, don't you want to be just like me?
  96. by the way, kuro5hin raised 37,000 dollars by turm5ric · · Score: 1

    that is stuff that matters

  97. MTBF by Tim+Locke · · Score: 1

    I wonder what the total Mean Time Between Failures is on all that equipment. It must be approaching zero.

    --
    *** On the Internet, no one knows you're using a VIC-20
  98. from the preview-of-2026-ipaq dept. by The+Pi-Guy · · Score: 2

    You mean the model H3760?

    --pi

    ... Erm, sorry. That's 3760000000. They release too many of these things. And, it only costs 1/8 of it's model number, like the rest do!

  99. Nope by cameldrv · · Score: 1

    For cracking codes, the NSA has computers that are far faster than this.

  100. Rejected by theolein · · Score: 1

    I submitted this two months ago on the 26th of April and it got rejected. I sometimes would like to know if it's only alcohol they use to select their stories or if other drugs are involved as well?