Slashdot Mirror


Top 500 Supercomputers Ranked

Shadow Wrought writes "The Register is reporting on (alternate ZDNet article) the latest list of the top 500 supercomputers in the world. Top of the list is the Earth Simulator Center in Yokohama, Japan, with a benchmark performance of 35.86 Tflop/s. HP and IBM claim 159 and 158 of the systems respectively. I wonder how many teraflops Deep Thought could have done?"

266 comments

  1. isn't the answer obvious? by sweeney37 · · Score: 5, Funny

    I wonder how many teraflops Deep Thought could have done?

    42.

    Mike

    1. Re:isn't the answer obvious? by MADCOWbeserk · · Score: 1

      We'll never know the question. Those strange apelike creatures broke the darn thing while drilling for oil.....

    2. Re:isn't the answer obvious? by glenebob · · Score: 1, Troll

      I *told* you, you weren't going to like it very much...

  2. atari by chimpo13 · · Score: 2, Funny

    Where does the Atari 2600 rank? First or second?

    1. Re:atari by Anonymous Coward · · Score: 2, Funny

      I want a cluster of 2600's running a high-res version of ET.

    2. Re:atari by forinti · · Score: 1

      first or second...to last

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

      I programmed the 2600. Hmm...You could have one processor/scan line....

  3. Deepthought by Duncan3 · · Score: 4, Insightful

    Since it was a chess computer, I'd say it could probably do about 0.000 FLOPS.

    --
    - Adam L. Beberg - The Cosm Project - http://www.mithral.com/
    1. Re:Deepthought by Anonymous Coward · · Score: 1, Informative

      WTF are you talking about? Deep Thought wasn't a chess computer, idiot. That was Deep Blue.

    2. Re:Deepthought by Distinguished+Hero · · Score: 1

      I think he was referring to the fictional computer from The Hitchhiker's Guide to the Galaxy after which the chess computer was named, not the chess computer itself.

      --
      Uttering logically derived and empirically supported truths to the disciples of the orthodox establishment.
    3. Re:Deepthought by Copid · · Score: 1

      Actually, before Deep Blue's time (and, apparently, before yours), there was a chess computer project called Deep Thought. I think that's what Duncan3 was referring to.

      --
      An interesting anagram of "BANACH TARSKI" is "BANACH TARSKI BANACH TARSKI"
    4. Re:Deepthought by sTalking_Goat · · Score: 0
      Since it was a chess computer, I'd say it could probably do about 0.000 FLOPS.

      The original Deep Thought dumb-ass.

      While this is /. and its a given that you're not going to RTFA. The least you could do if follow the links explaining some of the not so vague references in the summary.

      Or alternately read Hitchikers Guide to the Galaxy. A defenite modern classic, if such a thing exists.

      --

      My days of not taking you seriously are certainly coming to a middle...

    5. Re:Deepthought by Anonymous Coward · · Score: 0

      DUCK!!!
      whoooooooooooooooooosh!
      What the fuck was that?

    6. Re:Deepthought by tarquin_fim_bim · · Score: 1

      By crikey Sir, I do beleive that was the original post, cleverly positioned at about neck height to fool the unwary.

    7. Re:Deepthought by lostchicken · · Score: 2, Informative

      For those who don't get it, chess computers don't really deal with floating point math, but are dependent upon integer calculations. So their FLOPS scores are very low while their MIPS scores are outstanding.

      --
      -twb
    8. Re:Deepthought by Blue23 · · Score: 1

      If I remember, Deep Thought's "magic" was in the special chess evaluation chips, which wouldn't of been able to contribute to any general purpose floating point calculations.

      Without those, it was just a high-end RS/6000 SP cluster. Nice, but not one of the most powerful supercomputers in the world.

      =Blue(23)

      --
      LITTLE GIRL: But which cookie will you eat FIRST? C. MONSTER: Me think you have misconception of cookie-eating process.
    9. Re:Deepthought by ralphus · · Score: 1
      But Deep Thought wasn't the fastest computer ever. It was the fastest computer at the time and it designed the fastest computer ever to find the question to the answer (42) to life the universe and everything.

      That computer was Earth. So the fastest fictional computer ever was Earth, and the fastest real super computers are used as earth simulators. Now isn't that cool?

      I miss Mr Adams.

      --
      Revolutions are never about freedom or justice. They're about who's going to be top dog. -- Kilgore Trout
    10. Re:Deepthought by LearningHard · · Score: 0
      Deep Thought was the computer in Hitchhiker's Guide to the Galaxy that computed the answer to Life, The Universe, and Everything.

      Deep Blue is what you are thinking about.

    11. Re:Deepthought by FreakyDeaky · · Score: 1

      Deep thought was a chess computer before deep blue came around.

    12. Re:Deepthought by Anonymous Coward · · Score: 0

      No, Deep Thought (the chess computer), was named for the where the LINK leads to... dolt. The H2G2 computer.

  4. Sigh by Qweezle · · Score: 5, Funny

    I guess for the third straight year my ultrafast 17 kiloflop TI-83 calculator didn't make the list... :sigh:

    1. Re:Sigh by oO0OoO0Oo · · Score: 1

      What's that weird smell?

      Oh, that's just your TI-83 overheating and melting.

      --
      We Are Familiar With Elephants By Virtue Of Their Size.
    2. Re:Sigh by intermodal · · Score: 1

      yeah..my wife was disappointed her Athlon 1800+ XP didn't make it, even though I'd told her it was pretty fast when I built it for her.

      --
      In SOVIET RUSSIA... erm...NSA AMERICA, the Internet logs onto YOU!
    3. Re:Sigh by Anonymous Coward · · Score: 0

      I hope you have that bad boy supercooled!!!!

  5. Apple Macs by rebeka+thomas · · Score: 4, Funny

    I wonder when the first mac user will claim a G5 should be on the list Typical bigoted zealots.

    --
    RST
    1. Re:Apple Macs by rebeka+thomas · · Score: 1

      I was modded down within a minute of posting that comment

      Kind of proves my point doesn't it

      --
      RST
    2. Re:Apple Macs by Anonymous Coward · · Score: 0

      Flamebait? Is the truth flamebait?

    3. Re:Apple Macs by Anonymous Coward · · Score: 0

      set your threshold to -1, AC got in there third

    4. Re:Apple Macs by stoolpigeon · · Score: 4, Funny

      First- here is the answer.

      second: being modded down is the confirmation of your post- if they hadn't you would have been wrong. It was worth it. If I hadn't just burned my last point i'd have modded you back up.

      --
      It's hard to believe that's how Micronians are made. Why don't we see it right now by having you both kiss one another?
    5. Re:Apple Macs by Talisein · · Score: 0, Offtopic

      "Typical bigoted zealots"

      Not flamebait?

      --
      "The right to do something does not mean doing it is right." William Safire
    6. Re:Apple Macs by ahector · · Score: 1

      I think it was the "bigoted zealots" comment that got you a flamebait mod. Whether or not your point is valid, that makes sense. And hey, no worries! Someone just modded you up to 1 Funny! :-P

      --
      sig
    7. Re:Apple Macs by Anonymous Coward · · Score: 0

      No, you got modded down for pretending to be a girl. Just wait til the nmap guy hax0rs you in revenge.

    8. Re:Apple Macs by Anonymous Coward · · Score: 1, Funny

      Typical bigoted zealots.

      Kindly look in the mirror, sir.

    9. Re:Apple Macs by nightznoe · · Score: 0, Offtopic

      Somebody plays VGAP :)

    10. Re:Apple Macs by Rob+Bos · · Score: 0, Offtopic

      wow, a VGA Planets reference on Slashdot. :)

    11. Re:Apple Macs by Anonymous Coward · · Score: 0

      > on the list Typical bigoted zealots.

      Where can I find that list?

    12. Re:Apple Macs by coolmacdude · · Score: 1

      Yes but at least we now have the speed to back up our bigotry.

      --

      -You may license this sig for only $6.99.
  6. Not hosted on one... by Anonymous Coward · · Score: 5, Funny

    My thinking is that the list of top 500 supercomputers isn't hosted on such a machine...

    1. Re:Not hosted on one... by ehiris · · Score: 1

      I'd be more concerned if they'd rate supernetworks.

    2. Re:Not hosted on one... by PsykhoKiwi · · Score: 1

      I have to wonder where they get their rankings from.

      I work at the University of Manchester in Manchester Computing. Our Cray (http://www.csar.cfs.ac.uk/turing/) is listed at position 120 with stats from 2000 which are out of date. It is also listed as our most powerful machine when it isn't. We have another machine which is far more powerful now (http://www.csar.cfs.ac.uk/green/) It is supposed to be one of the most powerful in Europe and has been upgraded several times in the last year alone.

      Then again, all the stats on the CSAR website are well out of date, for example Green is twice the size now it was when that photo was taken and there are a load of new machines in the machine room which might stand a chance at getting on the list aswell.

      I sent some updated stats last time this list was published but they paid no attention...

      --
      Just remember that if the world didn't suck we'd all fall off.
    3. Re:Not hosted on one... by Sique · · Score: 1

      Then why don't you submit your results?

      --
      .sig: Sique *sigh*
    4. Re:Not hosted on one... by PsykhoKiwi · · Score: 1

      Like I said in the last line, the last time they published this list I *did* send in some updated stats. I work for COS, not CSAR so I don't have the current stats and I don't see the point in going and getting the up-to-date stats if they're just going to ignore me again.

      --
      Just remember that if the world didn't suck we'd all fall off.
  7. How dare you! by JVert · · Score: 5, Funny

    I'm posting from a duron right now you insensitive clod!

    1. Re:How dare you! by Anonymous Coward · · Score: 0

      Who started this whole "insensitive clod!" tradition?

    2. Re:How dare you! by Anonymous Coward · · Score: 0

      Imagine a bewolf cluster of insenitive clod jokes... ...oh

    3. Re:How dare you! by Unregistered · · Score: 1

      Could Be Worse. Yes, i am actually posting form one. I can't wait to get back to civilization.

  8. Deep Thought by nother_nix_hacker · · Score: 3, Funny

    I wonder how many teraflops Deep Thought could have done?"

    Who cares, I wonder what the fps in quake 3 would be! :)

    1. Re:Deep Thought by Anonymous Coward · · Score: 0

      >I wonder how many teraflops Deep Thought could have done?
      Who cares, I wonder what the fps in quake 3 would be! :)


      Maybe Quake 3, but it still probably can't handle Doom III.

    2. Re:Deep Thought by Anonymous Coward · · Score: 0

      it could probably exceed 1fps in mesa3d!

      i feel the need: the need for speed

    3. Re:Deep Thought by gnovos · · Score: 3, Funny

      I wonder how many teraflops Deep Thought could have done?"

      Who cares, I wonder what the fps in quake 3 would be! :)


      You wouldn't like the result...

      Deep thought, I want to play qua....

      Done! You would have gotten 32 frags and your friend would have gotten 38. You would lose. Would you like to play again? No, don't answer that, I already know. DONE! This time you would have beaten your friend by 4 frags.

      --
      "Your superior intellect is no match for our puny weapons!"
  9. Imagine a beow. . . . by romec · · Score: 2, Funny

    Well someone was going to say it. . .
    I think that they should add their web server as an honorable mention because I can still connect to it (it took a while) despite the /. effect

  10. Re:They would go back in time by infinite9 · · Score: 5, Funny

    What if there were a beowulf cluster of the top 500 supercomputers?
    You'd still be modded down.

    --
    Disconnect your television. Do your own research. Draw your own conclusions. They're probably lying. Don't be a sheep.
  11. Ironically.... by Silvers · · Score: 1

    The Top500 server was slowed to a grinding halt of php time-outs by a number of measly desktops.

  12. Here's a thought by coolmacdude · · Score: 3, Interesting

    I bet if I buy 10 new Dual 2 Ghz G5s and cluster them I could make that list.

    --

    -You may license this sig for only $6.99.
    1. Re:Here's a thought by NerdSlayer · · Score: 1

      Nope, not even close.

    2. Re:Here's a thought by Anonymous Coward · · Score: 1, Informative

      With the lowest one on the list still weighin in at 245 GFlops, assuming each cycle could process 1 floating point operation (usually not a safe assumption) puts it at a minimum of 123 of those suckers. Start saving those pennies.

    3. Re:Here's a thought by agent+dero · · Score: 0, Troll

      As soon as you buy those 10 new G5's, i'll be glad to steal them from you.

      --
      Error 407 - No creative sig found
    4. Re:Here's a thought by JDWTopGuy · · Score: 1

      Start saving those pennies.

      I am, I am!

      --
      Ron Paul 2012
    5. Re:Here's a thought by coult · · Score: 2, Informative

      Actually the G5 has dual floating point units, which means it can do 2 double-precision floating point ops per cycle (maximum)...and technically speaking the altivec can do 4 single-precision floating point ops per second, which would allow for a theoretical peak of 16 GFlops per machine. Of course the actual performance would be lower, especially in a cluster. I don't think you'd need 123 machines though.

      --

      All is Number -Pythagoras.

    6. Re:Here's a thought by coult · · Score: 1

      4 single-precision floating point ops PER CYCLE, not per second (oops)

      --

      All is Number -Pythagoras.

    7. Re:Here's a thought by coolmacdude · · Score: 1

      A haven't seen a Gfps estimate on the G5s yet, but I know the Dual 1 Ghz G4 was 15 Gfps. They have said that the new machines are at least twice as fast so that would put it at 30 Gfps. Now even assuming a typical 20-30% falloff for actual performance, it would still be at most a few dozen, not 123.

      --

      -You may license this sig for only $6.99.
    8. Re:Here's a thought by imsabbel · · Score: 1

      But thats an APPLE estimation
      This while working with larg array the locality of the data is very low. Mostly you do MACs (not apple Macs, but MultiplyAccumulates). It is true that you can get VERY high theoretical FLOP ratings using SIMD, but considere: In the end, you still have to write a result to memory and read a new input value.
      30 Gfps would need 240 GB/s memory bandwith.
      This arent your wheatstone benchmarks where all code and data fits into l1 cache.

      --
      HI O WISE PRINCE. WHT TOOK U SO DAM LONG?
    9. Re:Here's a thought by khuber · · Score: 1

      Subtracting out 29 Gfps for Quartz, that still leaves a decent 1 GFps.

    10. Re:Here's a thought by coolmacdude · · Score: 1

      In 10.1 you might be right. However with Quartz extreme most of the graphics processing no longer requires the CPU.

      --

      -You may license this sig for only $6.99.
    11. Re:Here's a thought by ActiveSX · · Score: 1

      BZZT, wrong. The only rendering pushed out to the GPU by QE is the window compositor. Everything else is done in software.

    12. Re:Here's a thought by Anonymous Coward · · Score: 0

      4 single-precision floating point ops PER CYCLE, not per second (oops)

      No, no, I've used a Mac before, and can confirm that you were right the first time.

    13. Re:Here's a thought by Anonymous Coward · · Score: 0

      Nope, not even close. Here's why... lets say the CPU can execute 4 double precision instructions per cycle like the Itanium does (I'm not sure that the G5's cpu can, maybe it can only do two). So then a 2 GHz CPU can achive 8 GFLOPS as a theoretical peak. A dual CPU box would have a tpeak of 16 GFLOPS.

      So, 10 of these boxes have a tpeak of 160 GFLOPS. Multiply that by 0.6 to get a number close to real world performance and you get 96. Note that the 0.6 ratio is difficult to achieve for #1) clusters without low latency networking such as Myrinet, and #2) dual cpu systems which need to share resources between the processors. If the G5 can only do 2 instructions per cycle, the numbers decrease to 48 and 80 for a 10 box cluster.

      Anyway, the lowest score on the top500 list is 245.1 with a tpeak of 384.

  13. Slashdotted already? by Anonymous Coward · · Score: 0, Redundant

    Obviously the list isn't hosted on one of those computers.

  14. The Top 10 by Anonymous Coward · · Score: 5, Informative

    The site may rank Supercomputers, but obviously doesn't run on one :) It's already chugging. Here's a straight, unformatted, copy and paste of the top 10:

    1 NEC
    Earth-Simulator/ 5120 35860.00
    40960.00 Earth Simulator Center
    Japan/2002
    2 Hewlett-Packard
    ASCI Q - AlphaServer SC ES45/1.25 GHz/ 8192 13880.00
    20480.00 Los Alamos National Laboratory
    USA/2002
    3 Linux Networx
    MCR Linux Cluster Xeon 2.4 GHz - Quadrics/ 2304 7634.00
    11060.00 Lawrence Livermore National Laboratory
    USA/2002
    4 IBM
    ASCI White, SP Power3 375 MHz/ 8192 7304.00
    12288.00 Lawrence Livermore National Laboratory
    USA/2000
    5 IBM
    SP Power3 375 MHz 16 way/ 6656 7304.00
    9984.00 NERSC/LBNL
    USA/2002
    6 IBM
    xSeries Cluster Xeon 2.4 GHz - Quadrics/ 1920 6586.00
    9216.00 Lawrence Livermore National Laboratory
    USA/2003
    7 Fujitsu
    PRIMEPOWER HPC2500 (1.3 GHz)/ 2304 5406.00
    11980.00 National Aerospace Laboratory of Japan
    Japan/2002
    8 Hewlett-Packard
    rx2600 Itanium2 1 GHz Cluster - Quadrics/ 1540 4881.00
    6160.00 Pacific Northwest National Laboratory
    USA/2003
    9 Hewlett-Packard
    AlphaServer SC ES45/1 GHz/ 3016 4463.00
    6032.00 Pittsburgh Supercomputing Center
    USA/2001
    10 Hewlett-Packard
    AlphaServer SC ES45/1 GHz/ 2560 3980.00
    5120.00 Commissariat a l'Energie Atomique (CEA)
    France/2001

    1. Re:The Top 10 by Aardpig · · Score: 2, Informative

      From this list, would I be correct in thinking that no Microsoft products (i.e., operating systems) run on these ultra high-end machines? Or, to paraphrase what I really mean (and in the interests of honesty), does Windows suck like a Dyson when it comes to High-Performance Computing?

      --
      Tubal-Cain smokes the white owl.
    2. Re:The Top 10 by $exyNerdie · · Score: 1

      From this list, would I be correct in thinking that no Microsoft products (i.e., operating systems) run on these ultra high-end machines? Or, to paraphrase what I really mean (and in the interests of honesty), does Windows suck like a Dyson when it comes to High-Performance Computing? It does mostly but here's an exception: Making a rare appearance: a Windows-based system, in the form of a cluster of Dell Computer servers with 384 Xeon processors. It was built at the Cornell Theory Center and is ranked No. 50 on the list

    3. Re:The Top 10 by TheViffer · · Score: 2, Informative

      Microsoft PR

      Because of their ground-breaking work with Velocity, CTC was recognized in Washington by ComputerWorld and the Smithsonian American History Museum and was made part of the Smithsonianâ(TM)s permanent research collection. Velocity was also named to the list of the top 500 most powerful computers in the world. This was a watershed event since it was the first Windows 2000-based system to obtain this ranking and one of only two Windows-based systems to place on the list.

      So in the list thier is another.

      --
      -- Knowing too much can get you killed, but knowing who knows too much can make you rich.
    4. Re:The Top 10 by sTalking_Goat · · Score: 0
      Wow 3 of the top six reside at Lawerence Livermore.

      Makes me want to rethink leaving the whole high energy physics thing....

      --

      My days of not taking you seriously are certainly coming to a middle...

    5. Re:The Top 10 by shfted! · · Score: 2, Interesting

      There is a notible lack of ASCI systems built with Power4 chips. I wonder how long until IBM will build one with them?

      --
      He who laughs last is stuck in a time dilation bubble.
    6. Re:The Top 10 by Strog · · Score: 1

      There are several Power3 setups in the list including #4 and #5. Not bad for 375Mhz chips and you really have to wonder what the newer chips are capable of.

    7. Re:The Top 10 by perky · · Score: 1

      number 57.

      Self-made
      Pegasus P4 Xeon Cluster 2.2/2.4/2.8 GHz - Giganet - MSWindows/ 400

      --
      "The new wave is not value-added; it's garbage-subtracted" - Esther Dyson, Dec 1994
    8. Re:The Top 10 by redgren · · Score: 1

      Power4 ASCI will be delivered by end of the year. Will be most impressive, as are all of the ASCI systems.

  15. Is this really news? by Anonymous Coward · · Score: 0

    Is the top500 site going to get posted on /. everytime it is updated?

    1. Re:Is this really news? by Oldskooldave · · Score: 1

      ........probably

  16. Nebraska doesn't suck by rk2z · · Score: 1

    # 200 Prairiefire AMD 1.8 GHz - Myrinet/ 256 University of Nebraska UNL doesn't suck, WOOT!

    --
    This is a sig, there are many like it, but this is mine.
    1. Re:Nebraska doesn't suck by TheViffer · · Score: 1, Funny

      No .. generally it blows, it is Iowa that sucks.

      --
      -- Knowing too much can get you killed, but knowing who knows too much can make you rich.
    2. Re:Nebraska doesn't suck by redheaded_stepchild · · Score: 1

      It's awfully windy in Kansas, Oklahoma must suck too.

      --
      Don't use the Troll mod just because you disagree with me.
    3. Re:Nebraska doesn't suck by Anonymous Coward · · Score: 0

      Yea, Nebraska sucks, just live with it.
      #137 - Xeon 2.4 Cluster Myrinet - University of Oklahoma - Fastest in the Big XII!
      Boomer Sooner!

  17. Re:They would go back in time by ianjk · · Score: 1

    The world would implode and suck the rest of the universe into the darkness, until god's pager goes off at a weekend golf retreat signaling that his 8 Billion year uptime just went to shit. He runs home, hits 'ctrl+alt+delete' and reboots. It all begins again.

    I really need to stop abusing the free pop machine @ work.

  18. Alternate site by Dynamoo · · Score: 4, Funny
    A similar presentation of the data can be found here cuz the main one has just gotten Slashdotted.

    Never mind Teraflops, we should have a measure of web server load called "Slashdots".

    --
    Never email donotemail@WeAreSpammers.com
    1. Re:Alternate site by Anonymous Coward · · Score: 0
      How is Sun listed with F15k with 144 procs?

      Max on F15k is 18*4 + 17*2 or 106 procs which is a very IO limited configuration.

      Must be a cluster of two machines @72 procs each.

    2. Re:Alternate site by stevenp · · Score: 1

      >> Never mind Teraflops, we should have a measure of web server load called "Slashdots".

      One Slashdot is too much for an average site.

      I think most sites will be measured in mini- and micro-slashdots. The amateur sites could be somewhere in the pico-slashdot range.

      A truly heavy-bandwith-distributed-load-balanced-admin-sch izofrenied site cound eventually make 1-2 Slashdots

      As the small values are hard to evaluate (how do you distinguish between 1 and 5 pico-slashdots?) I beleave an alternative measurement could take place - Time-To-Slashdot (TTS for short) measured in seconds. It will be the average time that the site survives after a link to it is posted on the front page of Slashdot.

      - I know you are looking at me

  19. Woo hoo! by Anonymous Coward · · Score: 0

    We're numbers 39 and 189!

  20. I just bought a Macintosh by Anonymous Coward · · Score: 1, Funny

    Why isn't it listed as a supercomputer? Jobs lied to me!

  21. Imagine by News+for+nerds · · Score: 1

    Beowulf... no, imagine that one of those computers will come on your desktop or something in, say, 15 years. (2^15 = 32768)

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

      Well, not to be pedantic, but technically speaking, Moore's law predicts a 2 x increase every 18 months, so 15 years = 10 * 18 months.

      2^10=1024.

      Call back in 22.5 years.

  22. Phaeton Sez by Anonymous Coward · · Score: 0

    Damn.. Slashdotted already.

    This isn't hosted on a SuperComputermaschine, is it?

  23. Can't be by phorm · · Score: 0, Redundant

    Because if that were the answer - then 5 minute prior to you posting it the earth would have been destroyed to make well for an interstellar highway.

    Geeze, I just finish reading hitchhiker and realize just how many quotes I've been missing out on

    1. Re:Can't be by Xerithane · · Score: 0, Redundant

      Because if that were the answer - then 5 minute prior to you posting it the earth would have been destroyed to make well for an interstellar highway.

      You need to read it again, mate. The earth was destroyed while calculating the question. 42 was just the answer.

      --
      Dacels Jewelers can't be trusted.
    2. Re:Can't be by Surak · · Score: 1, Redundant

      The question is "How many teraflops does Deep Th...."

      *KABLOOM* *the Earth is suddenly destroyed by a Vogon Destructor fleet to make way for a new hyperspace bypass*

    3. Re:Can't be by phorm · · Score: 0, Redundant

      It's the answer to the question of what the question to the answer to the question of the meaning of life is.

      *ouch*
      I think that Adams must have been a programmer at one point... he has this knack for literary loops...

    4. Re:Can't be by FreakyDeaky · · Score: 1

      I wonder What percent of /. readers have read Hitchhiker's guide I'm guessing around 80%.

  24. American Dominance in Supercomputers by reporter · · Score: 4, Interesting
    Before anyone says that Japan is "ahead" of the USA again, let me quell any doubts about the superiority of American computer technology. The technology to build a HPC (high-performance computing) computer is identical to the technology to build a commercial computer. American companies have been increasingly focused on building commercial computers because that is where the profits are -- especially after the end of the cold war.

    The NEC Earth Simulator is really just a different optimization point in the computer-design space. Huge amounts of bandwidth to memory and specialized vector-processing units tied to the processor core. The VLSI technology that NEC used to build these system is readily employed by Intel and IBM. So, if the latter companies wanted to build the world's fastest HPC computer, they could.

    The 21st century is not PaxAsia. It is PaxAmericana. The hordes of immigrants flooding into this country to get the hell out of Asia should have been a big hint.

    1. Re:American Dominance in Supercomputers by Anonymous Coward · · Score: 1, Interesting

      I wasn't thinking that at all, I was just thinking "Oooh look at the big shiney supercomputers".

      Does the fact that you felt the need to leap to the defence of the US of A's supercomputing programs before anyone questioned them not show that you are a little paranoid and pessimistic about how well they are doing ???

      To be honest where they are located is the last bit of information I looked at.

    2. Re:American Dominance in Supercomputers by agent+dero · · Score: 5, Interesting

      The difference in manufacturing in Japan and the US are another extension of the cultural differences. (I believe)

      For instance, in Japan many manufacturers are creating smaller devices, something that hits the market has to be smaller and more "efficent" (not costing lots of Yen in overhead) This is something typical of the Japanese culture. For example, you won't see some Japanese Guy driving around in an Escalade with his girlfriend, "just cause" You'd more likely find some Japanese Guy driving around in a Honda Hybrid car, cause it's cheaper to run than most cars, even though the price sucks.

      Meanwhile, in the States, people want more POWER! They want the big ass SUV that crushes other cars, and small animals. The same goes with the computers. Something big and fast, regardless of the power consumption or general overhead of the machine.

      Therefore, the target markets in the two countries are much different, so the products of the two will also be much different.

      --
      Error 407 - No creative sig found
    3. Re:American Dominance in Supercomputers by Anonymous Coward · · Score: 0

      Nice troll. Sounds like someone's got some personal issues.

    4. Re:American Dominance in Supercomputers by Anonymous Coward · · Score: 0

      I dont know about the rest of us Americans, but I love my hybrid Toyota Prius. :)

    5. Re:American Dominance in Supercomputers by Arti · · Score: 1

      "Pax" is a funny word for the [x] Americana we are about to experience. Also I'm not sure if PaxAmericana is a legitimate portmanteauword.

    6. Re:American Dominance in Supercomputers by Otter · · Score: 1
      This is something typical of the Japanese culture. For example, you won't see some Japanese Guy driving around in an Escalade with his girlfriend, "just cause"...

      Well, you do sometimes, but those drivers tend to have tattoos, perms and missing fingers, and you probably want to stay the hell away from them.

    7. Re:American Dominance in Supercomputers by Anonymous Coward · · Score: 0

      Sounds like you have a bad case of the "I hate America"'s. Been watching too much of the liberal controlled media, lately? Fucking Frenchie.

    8. Re:American Dominance in Supercomputers by Eminor · · Score: 1

      Yes, America is better at everything!

    9. Re:American Dominance in Supercomputers by Anonymous Coward · · Score: 0

      Ha! Dumbass, if it wasn't for the French backed "War of Independence" you'd still be part of the British Empire just like those dozey Canooks.

    10. Re:American Dominance in Supercomputers by Anonymous Coward · · Score: 0

      Actually, although the Americans are a little bit obsessed with 'biggest and best', the Japanese are much more obsessed with this.

      This is why they built the largest battleship ever made (the Yamato). This is why their kids watch cartoons about giant robots. This also explains why they built the fastest supercomputer in the world.

      (And don't ask about Sumo wrestling!)

    11. Re:American Dominance in Supercomputers by Anonymous Coward · · Score: 0

      Why does who hate America ?

      I didn't see any hate ...

      The first post just sounded a bit like "Well your computer might be bigger than our computer, but we could so that if we wanted to ... we just don't want to"

    12. Re:American Dominance in Supercomputers by Anonymous Coward · · Score: 1, Funny

      Including being humble.

    13. Re:American Dominance in Supercomputers by Anonymous Coward · · Score: 0

      Wow! You're just a weirdo!

    14. Re:American Dominance in Supercomputers by Aardpig · · Score: 1

      The 21st century is not PaxAsia. It is PaxAmericana

      I think "American Global Dominance" would be a better term than "Pax Americana". Recalling my Latin, "Pax" means peace, and you just have to look at Afghanistan and Iraq to see the inappropriateness of the term "peace" in this context.

      --
      Tubal-Cain smokes the white owl.
    15. Re:American Dominance in Supercomputers by Anonymous Coward · · Score: 0

      So you're telling me that asians are proportionately over-compensating for their tiny noodle-dicks?

    16. Re:American Dominance in Supercomputers by Anonymous Coward · · Score: 0

      I wish you'd left out that last paragraph...

    17. Re:American Dominance in Supercomputers by netsharc · · Score: 2, Insightful

      The 21st century is not PaxAsia. It is PaxAmericana. The hordes of immigrants flooding into this country to get the hell out of Asia should have been a big hint.

      Gee whiz, what a grand conclusion from a simplistic argument. I wonder why people still find USA attractive, I come from the world's largest Muslim country (not Muslim myself), and I wouldn't want to go to a country where I would immediately be seen as a suspect terrorist and where your Agent Smiths can arrest me for just because I'm foreign. I don't even think they'll let me in if I wanted to visit the goddamn country. The same fate is suffered by a lot of students and scientists who study in the US and had helped the US to be the great technological nation that it currently is. Here's the article which mentions that fact, but I don't think you'd be able to read it - it's in German and you don't seem like someone who'd learn a foreign language.

      That seems to be only a small, unimportant part of the US's big problem. Afghanistan is going to hell again, Iraq is still in chaos, and billions of your taxpayer money is being used in trying to stop that chaos, chaos which Bush created in hopes of rewards that he isn't going to share with you, because they're all going into his and his friend's and family's pockets.

      Yea, vote Republican!

      --
      What time is it/will be over there? Check with my iPhone app!
    18. Re:American Dominance in Supercomputers by cly · · Score: 1

      I don't know if you are just trolling or not, but I'll say something on this.

      Horst Simon (director of NERSC) gave a talk in our school a few months ago and talked about the difference in supercomputing in Japan and US.

      Basically what he said is, US companies have no incentive to build hardware for supercomputing - the market is not big enough, and the government does not provide the support (although the recent DoE HPCS initiative might change this). Meanwhile, Japan government put in a lot of money and effort in supercomputing - Earth Simulator is basically a government effort down to the hardware design stage.

      'To regain the leadership', he said, they are working with IBM now on the architecture of the next Power series architecture to make it more scientific computing friendly.

      It's not about the technology - both Japan and US are capable of building good supercomputers. And it's not clear that vector processing is really good for scientific supercomputing anyway (one example - earth simulator, won't prove that it's good).

      It's more about the market, and who is willing to put more money in.

    19. Re:American Dominance in Supercomputers by Anonymous Coward · · Score: 0

      America MUST BE better at everything, isn't it?

    20. Re:American Dominance in Supercomputers by Anonymous Coward · · Score: 0

      Computing speed is the top priority in some numerical simulations such as global environmental study which is critical to everyone on the earth. It is exactly why the Earth Simulater was constructed. Maybe US government and people are not interested in such issues more than nuclear weapons.

    21. Re:American Dominance in Supercomputers by 2short · · Score: 1

      So now recall your history, and decide if he's making a reference to the "Pax Romanus". Yes, of course it literally meant "Roman Peace", but you only had to look at Britain, Gaul, etc. to see that "Globulus Dominatus Romanus" would have been a better term.

      There's this thing called "irony". You may wish to look it up, lest you critisize those who are making your very point, but more effectively.

    22. Re:American Dominance in Supercomputers by Explo · · Score: 1

      The 21st century is not PaxAsia. It is PaxAmericana. The hordes of immigrants flooding into this country to get the hell out of Asia should have been a big hint.

      Only about 97 years of this century left - are you sure that your conclusion isn't somewhat bold? ;)

      --
      Everyone who makes generalizations should be shot.
    23. Re:American Dominance in Supercomputers by Aardpig · · Score: 1

      There's this thing called "irony". You may wish to look it up, lest you critisize those who are making your very point, but more effectively.

      There's this thing called "spelling". You may wish to look it up, before you spell "criticize" incorrectly.

      --
      Tubal-Cain smokes the white owl.
    24. Re:American Dominance in Supercomputers by Anonymous Coward · · Score: 0

      There's this thing called a life, you may wish to get one before you waste it spell checking slashdot posts.

  25. I must be looking at too many porn/warez sites by zapp · · Score: 4, Funny

    Anyone else get the instant urge to close the window (thinking it was a popup) when they saw the host www.top500.org?

    --
    no comment
  26. Changes? by MightyPalm · · Score: 3, Interesting

    I don't see any major changes in this list compared to the one that has been up for almost 6 months. Only a couple of the computers on top25 has been been build this year. I'm certainly looking forward to see some new top placements in the near future (anyone know of something which might appear soon?). btw. the server isn't too fast, even when it's not slashdotted.

    --
    Digital Evolution - Unregulated knowledge is pornography
    1. Re:Changes? by Anonymous Coward · · Score: 0

      I did too. I guess I sapent way too much time in the warez scene when I was younger.

  27. Some thoughts... by anzha · · Score: 5, Interesting

    Interesting to note is that #3, #6, and #8 are all linux clusters. All three of which are at Livermore.

    Cray's X1 also debuted, but it was much lower @112. However, it ought to be noted, that the examples out so far are only 60 processors at tops. As soon as the money gets ponied up, prolly at ORNL, they'll be waaaay up towards the top. My guess is, if all goes as planned, they'll be at #15 by year's end.

    What I find exciting these days is actually the High Productivity Computing Systems Effort, the Blue Planet or Blue Gene. These are a little ways off from being on the Top500 list yet though. :D

    I do wish there were more SC companies doing hardware development in the US. I love Cray, but a single vendor smacks of eggs in one backet syndrome...So, geeks, if ya wanna start a startup with a design, go for it...Betcha the NSA (aka Cthuhlu of HPC) would be happy to sponsor ya...;)

    --
    Do you know why the road less traveled by is littered with the bones of the unwary?
    1. Re:Some thoughts... by Surak · · Score: 1

      Cray's X1 also debuted, but it was much lower @112

      Yeah, but can you imagine a B...oh never mind....

    2. Re:Some thoughts... by salimma · · Score: 1
      Betcha the NSA (aka Cthuhlu of HPC) would be happy to sponsor ya...;)

      Of course, we probably would not hear about it until someone else reinvented something and the NSA said oh, by the way, we did this 10 years ago...
      --
      Michel
      Fedora Project Contribut
    3. Re:Some thoughts... by Noren · · Score: 1
      While #8 is a linux cluster, it is not at Livermore. It's at PNL, up in Washington State.

      ASCI White (#4) is at LLNL, but it is not a linux cluster- perhaps that contributed to the confusion.

    4. Re:Some thoughts... by anzha · · Score: 1

      Brain fart...my apologies. d'oh. I glanced a wee bit too fast.

      --
      Do you know why the road less traveled by is littered with the bones of the unwary?
    5. Re:Some thoughts... by alcourt · · Score: 1

      Which brings up the real reason this list is worthless. We as Joe Q Public have little to no way of knowing what kind of supercomputers were delivered in secret or confidentially.

      It's a brag fest for those who buy supercomputers and don't care who knows their real compute power. I have a gut feeling that is actually a rather small set of people. If you read the About, they are based on surveys of manufacturers and users.

      Another note of interest is that the benchmark chosen is to solve a dense system of linear equations. Depending on what kind of problem you are buying supercomputer time for, that may not be the best benchmark. What if you are trying to solve problems based on very large sparse matrices and very fast random memory access is the defining bottleneck instead? Or another problem may require one to use integer math instead of floating point math? Such things can radically change the performance of a system. (Or to be really evil, what if your problem does not parallelize well at all? Suddenly a cluster may not be anywhere near as useful as it once was.)

      --
      "I may disagree with what you say, but I will defend unto the death your right to say it." -- Voltaire
  28. Yeah, but... by Anonymous Coward · · Score: 0

    yeah, but people are gonna mod him up as interesting... typical.

  29. We're on the edge by koh · · Score: 1, Troll

    Is this what slashdot evolved into ? A top-500 list of supercomputers decided upon using dubious benchmarks that are not representative of computing power ?

    Doesn't any of us remember that FLOPs, as MIPS, are Meaningless Indication of Processor Speed ?

    I'm feeling a little deceived actually...

    --
    Karma cannot be described by words alone.
    1. Re:We're on the edge by msgmonkey · · Score: 4, Informative

      Well it depends, when you're talking about your average desktop computer the quoted FLOPS or MIPS are usually useless because the supporting architecure does n't have bandwidth to supply the processor the data at that speed, they are normally based on data that is in cache memory.

      Supercomputers are designed with high bandwidth in mind and thats why in general their FLOPS are taken with less of a pinch of salt.

    2. Re:We're on the edge by Anonymous Coward · · Score: 0

      > is this what slashdot evolved into?

      Evolved?

      Slashdot?

      Ahhhhhhhhahahahahaha!

      You must be new here.

    3. Re:We're on the edge by smitty45 · · Score: 4, Informative

      If you don't have any idea of why FLOPs aren't meaningless, then you haven't ever run a program/problem/simulation on these machines able to put the numbers into context for you.

      I spent 4 years running dynamic finite element analysis simulations on alot of the kinds of these parallel monsters, and when FLOPs indicate numbers that reflect quite well the length of time it would take for a run to finish, you realize that benchmarks ARE useful, in the right context.

  30. when will we have clustering for small business? by civilengineer · · Score: 1

    How long is it before consulants and engineers run their simulations (like road traffic and passengers at airports) on small sized supercomputers? So, far it seems like single desktop is still the norm.

    --

    New year Resolution: Don't change sig this year
  31. The top 500 unclassified supercomputers by mikeophile · · Score: 4, Interesting

    What I want to see is the list amended to include the iron that agencies like the FBI, CIA, NSA, and less well known acronyms are using.

    1. Re:The top 500 unclassified supercomputers by haystor · · Score: 1

      Iron?

      Hah! Only amaeteurs use iron. The real agencies use any computer and send it back in time so the answer is already in the database.

      --
      t
    2. Re:The top 500 unclassified supercomputers by hawkstone · · Score: 4, Interesting

      #2 (Q at LANL) and #4 (White at LLNL), and others, are doing classified work, but they're still on this list. But you're pretty much right if you meant that these are all unclassified architectures.

      Interestingly, though, things the NSA has wouldn't be likely to show up on this list, as the benchmarks are suited towards MPP style machines. NSA is more likely to have vector machines than large numbers of scalar processors.

    3. Re:The top 500 unclassified supercomputers by heli0 · · Score: 1

      Well considering the Earth Simulator Project got ~$1Billion(US) to get going and the DOD gets $356Billion every year... I am sure that after 9/11 the NSA could write a blank check to put a system in place to break 32768-bit encryption.

      --
      Whenever the offence inspires less horror than the punishment, the rigour of penal law is obliged to give way...
    4. Re:The top 500 unclassified supercomputers by El · · Score: 1

      We could tell you -- but then we'd have to kill you.

      --

      "Freedom means freedom for everybody" -- Dick Cheney

    5. Re:The top 500 unclassified supercomputers by sysjkb · · Score: 4, Interesting

      An alternative list of supercomputers is (or perhaps "was") kept by Gunter at GAPCON. He ranks sites, as opposed to individual systems. Unfortunatly, the site seems to be down right now, but the ever helpful google cache (search for "gapcon supercomputer") put the NSA at #4, behind the Japan Marine Science Center, LLNL and LANL.

      12475.7 - (09-JAN-2002) [NSA]
      National Security Agency,Fort Meade,Maryland,US
      1) Cray X1-3/192 2457.6
      2) Cray T3E-1200E LC1900 2280
      3) Cray T3E-900 LC1324 1191.6
      4) SGI 2800/250-2304 1152
      5) HP SuperDome/552-512 1130.5
      6) Cray T3E-1350 LC800 1080
      7) SGI 3800/400-1064 851.2
      8) Cray T3E-1200E LC540 648
      9) Cray T3E-1200E LC540 648
      10) Cray T3E-1200E LC540 648
      11) Cray T3E-1200 LC404 484.8
      12) Cray T3E-1200 LC284 340.8

      I think much of his information comes out of press releases, leaks, submissions, etc. The record for the NSA hadn't been updated for a while, but this may give you an idea of what they might have been running not very long ago.

      The modern stuff is all well and good, but what's really fascinating is his list of the top computing sites in 1956. (Or search google for 1956 computing sites and click on the cache.) Here's the top two. Rating is in OPS:

      583733.3 - [ONR]
      Office of Naval Research,Arlington,Virginia,US
      1) MIT Whirlwind 1 500000
      2) ERA Atlas 2 83333.3
      3) ERA 1101 200
      4) ERA Atlas 1 200

      253787.8 - [MIT]
      MIT,Cambridge,Massachusetts,US
      1) MIT TX-0 166666.7
      2) MIT Whirlwind 2 45454.4
      3) IBM 704 41666.7

      Note that the supreme super-computer of that era, the Whirlwind, is quite a bit slower than your pocket calculator.

      Yours truly,
      Jeffrey Boulier

    6. Re:The top 500 unclassified supercomputers by Saeger · · Score: 1
      Silly wabbit, the gubbermint has been using reverse engineered alien computing technology since the 50s. It's true!

      These Top Secret quantum computers are hidden miles beneath the artic pole in the joint underground bunkers of the Illuminati, The High Jew Command, and the Grey Alien Galactic Council. It's all true! Really!

      This super-secret computer is primarily used to simulate our solar system in enough detail such that The Powers That Be can "guide" history by killing the right butterflies to maximize their power. It's all true! Sincerely! Check out the literature yourself!

      --

      --
      Power to the Peaceful
    7. Re:The top 500 unclassified supercomputers by Anonymous Coward · · Score: 0

      From the 2000 Federal Budget...
      Hybrid Technology
      Multithreaded (HTMT)
      architecture
      Researchers funded by DARPA, NASA, and the National Security Agency
      (NSA) are evaluating the feasibility of constructing a computing system capable of
      a sustained rate of 1015 floating point operations per second (one petaflop).
      Preliminary evaluations of conventional architectures and mainstream technologies
      indicate that such a system will necessitate a radical departure from current
      research. HTMT architecture would blend modified semiconductor technology
      with leading-edge hybrid technologies including superconducting technology,
      optical interconnects, high speed very large scale integration (VLSI) semiconductors,
      and magnetic storage technology configured to satisfy the architecture
      requirements. Fundamental drivers are multi-gigahertz speeds, exceptional bandwidths,
      and very large and cost-efficient memory size. If results from FY 1999 R&D
      are encouraging, researchers are expected to build major sections of a prototype
      beginning in FY 2000.

    8. Re:The top 500 unclassified supercomputers by joper90 · · Score: 1

      if you have a fast computer they can just walk in and takeover/use it. i'm not joking..

    9. Re:The top 500 unclassified supercomputers by Scorchio · · Score: 1

      The UK Government Communications Headquarters (GCHQ) are due to relocate to a shiny new building nicknamed "the doughnut", later this year. It's up the street from the house I'm renting at the mo. I remember a few months ago, reading an article either in the local newspaper or a brochure that came round explaining about the new building. One thing it mentioned was a large hall constructed to house the most powerful super computer outside of the US. Now, that kind of infers that the most powerful computer in the world is in the US, which the top 500 article disagrees with. Or maybe the list doesn't include some classified ones. Or, as the GCHQ claim was future tense, maybe they're expecting things to change soon. Whichever, I'm putting aside my roll of cat5 and ceasing work on the tunnel until I know what's what.

  32. Highest TFlops in one location by eericson · · Score: 2, Redundant

    My question to anyone out there that might now is what location has the highest computing resources. Looking at the list, it seems like Lawrence Livermore has at least 3 systems in the top 50, and more from there on down.

    Anyone have a toal available resources in one location list?

    -E2

    --
    The evil monkey commands you to dance.
  33. Re:They would go back in time by MoonBuggy · · Score: 1

    That's what slashdot itself is running on, how else could it survive what can only be described as a 24/7 superslashdotting ;-)

  34. linux on the supercomputers? by Anonymous Coward · · Score: 0

    Now how many of those run linux? SCO might be reading this too.

  35. Teraflops... errr.. by Anonymous Coward · · Score: 0

    So.. does this mean that this is done in real-time???

  36. If things were different in the 80's... by lscotte · · Score: 2, Insightful

    This list would probably have been dominated by elxsi, Connection Machines, and Crays, if things had been different.

    I never saw a live CM or Cray, but I did play on an elxsi, and it was a pretty hot system for it's time.

    --
    This post is licensed under the Creative Commons Attribution-Noncommercial-No Derivative Works 3.0 License.
    1. Re:If things were different in the 80's... by Dylan+Zimmerman · · Score: 1

      I got to see a Cray up close and personal a few times. They have one running the lights and such at Disney World (what a horrible waste), then there's one on display in a museum that I went to. I forget where it was. Possibly Los Alamos? Well, I never actually got to use one, but it amazed me that the wires sometimes had loops to make sure that all of the wires were the same length so that the bits from one part would hit another part at the same time.

      Anyway, I haven't seen a Connection Machine, either. I wonder how much it would cost to have one built using modern materials and technology. With the .09 micron manufacturing spec, far more than 16 of those cores should fit on a single chip. Of course, the vast majority of the volume of the CM-1 was taken up by the processor interconnects, so it can't be reduced too much without making the entire thing one giant chip. It takes a lot of wire to make a 12th dimensional cube.

      I love all those trolls that ask why Apple claims that their computers are "supercomputers". What standard do they use to determine supercomputer status, they ask. The fact is that they use the US governmentâ(TM)s standard. For quite some time, Macintosh computers were powerful enough to be considered weapons by the United States. It was illegal to export them to quite a few countries. When the prices on computers that fit that criterion dropped enough that the majority of computer sales were 'weapons grade computers', that law was changed.

  37. can i ... by civilengineer · · Score: 1

    check e-mail on those computers? :)

    --

    New year Resolution: Don't change sig this year
    1. Re:can i ... by Penguinshit · · Score: 1

      I did play the original "Dungeon" (anyone remember the Zork trilogy?) on the Cray at Ames Research in California back in the early 80s.

    2. Re:can i ... by Anonymous Coward · · Score: 0

      well lots of the linux clusters feature all the standard linux software so you could on them. I've found unused cluster nodes make a nice extension to my desktop pc :)

  38. Re:They would go back in time by sgups · · Score: 0, Offtopic

    God uses Windows? GODDAMN!!!

    --
    Democratic USA - Government of the corporations, by the Corporations, for the corporations.
  39. Re:when will we have clustering for small business by MightyPalm · · Score: 1

    In my experience, a lot of small businesses rent time on university supercomputers to do some heavy simulation. But i also know of several small companies that run their own clusters. With x86-based hardware at a relativly low cost, i think that is evolving to be much more common in the near future.

    --
    Digital Evolution - Unregulated knowledge is pornography
  40. My Rig by Anonymous Coward · · Score: 2, Funny

    Sheesh, I wonder why my rig didn't make it into the top 500 either?

    I've spend a lot on a window, neon lights, fancy cooling systems, and STICKERS.. so many that I should rank atleast 354 out of 500!! It even sounds fast from all the chassis fans!

    Maybe I need to go for the rear mounted spoiler to break onto the list??

    1. Re:My Rig by Anonymous Coward · · Score: 0

      You musta forgot to put on the TypeR sticker on the case.

    2. Re:My Rig by FreakyDeaky · · Score: 1

      just becuase your computer is flying around the room at 70 miles per hour, due to fan power, doesn't mean it's fast. put some more fans on it and you might be able to break the sound barrier.

  41. Idiots by Anonymous Coward · · Score: 0

    Yep, love them Mac-fags, huh? Say something blatantly ridiculously optimistic about the performance of a Mac, they'll mod it straight up. Easiest Karma-whoring on slashdot.

  42. Deep Thought by interstellar_donkey · · Score: 1

    Boy, am I out of it. I would've expected deep thought to not even register. I gotta start checking the horsepower of shell machines these days.

    (yes, I know what they meant, but for a minute I was scratching my head).

    --
    The Internet is generally stupid
  43. But Earth Simulator IS a BIG ASS machine! ;-) by PaulBu · · Score: 1

    Check THIS out:
    http://www.mext.go.jp/english/org/science/im age/35 _01.jpg
    (it is a three story building, and I guess it also has its own powerplant).

    For more pretty pix, of course:
    http://images.google.com/images?q=Earth+S imulator& ie=UTF-8&oe=UTF-8&hl=en

    Paul B.

    1. Re:But Earth Simulator IS a BIG ASS machine! ;-) by Yomlogs · · Score: 1

      That's just the prototype. Next year every Japanese phone will have one built-in.

  44. NONSENSE by Arti · · Score: 1

    If Deep Thought could perform more terraflops than the Earth Simulator Centre then the ultimate computer wouldn't have had to be built. Deep Thought could have emulated it.

    1. Re:NONSENSE by Negatyfus · · Score: 1

      Who says the builders of the Earth Simulator Centre knew of Deep Thought's existence? The poor inhabitant of this artificial planet didn't even know they were merely part of an elaborate computer program, let alone know what created it!

  45. The answer to Deep Thoughts speed! by Anonymous Coward · · Score: 4, Funny

    Given:
    Deep Thought spent 6,500,000 yrs designing the earth, which then failed to produce the desired results. ( Admittedly not due to design flaws but still a failure. )

    Conclusion:
    1.53846153e-6 Terra Flops per Year.

    1. Re:The answer to Deep Thoughts speed! by makapuf · · Score: 1

      NO, you're wrong. Mice designed earth, not Deep thought.

    2. Re:The answer to Deep Thoughts speed! by Usquebaugh · · Score: 1

      The mice paid for the earth to be designed.

      Slarty Bart Fast did a lovely job on the shoreline, glacier you know, wonan award he did.

    3. Re:The answer to Deep Thoughts speed! by JHelgie · · Score: 1

      From the book: "I speak of none other than the computer that is to come after me," intoned Deep Thought, his voice regaining its accustomed declamatory tones. "A computer whose merest operational parameters I am not worthy to calculate - and yet I will design it for you."

  46. compared to ST:TNG's Data? by Anonymous Coward · · Score: 0

    You know, Data's raw processing power doesn't seem that great anymore (60 TeraFlops...think I'm reading the page correctly...and 800 quadrillion bits of storage)

    http://users.telenet.be/cook/engineering/compute r. html

  47. PS2 by laserlights2000 · · Score: 0

    Seeing how its been slashdotted, how high does the PS2 cluster, if it is even in there, rank? If it is really high, I suppose we can imagine daisy chaining a bunch of modded xboxs.

    1. Re:PS2 by laserlights2000 · · Score: 0

      http://games.slashdot.org/article.pl?sid=03/05/23/ 2250224&mode=thread&tid=127&tid=186&tid=21 2

  48. Where is Dr. Evil?.. but seriously by V_IL_Len · · Score: 1

    Could some organization or individual (gates gov whomever) with a ton of money had a supercomputer of this size built but not allowed it to be listed or not told anyone that they built it or classified it as top top secret. I don't know what they would want it for but I there could be a couple out there.....

  49. What language? by Aardpig · · Score: 5, Interesting

    I'd find it fascinating to see a breakdown of the languages used to write the applications which are running on these beasties. High Performance Computing has rather different needs from a language than programs that are, say, focused on interaction with a user, or database access. I expect that languages which sit well with infrastructures such as MPI and Open MP would be well-favoured, and I wouldn't be surprised to find that FORTRAN puts in a respectable showing.

    And before I'm bombarded with comments pointing out how ancient FORTRAN is, it's worth remembering that FORTRAN is still an evolving language; the last updade came out in 1997/98, and the new FORTRAN 200x should be arriving within the next year or so. In my experience of a number of languages, I've found that FORTRAN still continues to excel at numerical efficiency and portability, and I hope these selling points continue to be a feature of the new standard. Of course, I wouldn't want to write a compiler in FORTRAN, but for stuff like computational fluid dynamics, it still rocks; and those aspects of FORTRAN 77 which made it awkward (such as lack of dynamic array allocation) are fortunately a thing of the distant past.

    --
    Tubal-Cain smokes the white owl.
    1. Re:What language? by Anonymous Coward · · Score: 0

      Do they have open source compilers for the more recent FORTRAN incarnations yet? I saw GNU was working on a FORTRAN-95 compiler.

    2. Re:What language? by Aardpig · · Score: 1

      AFAIK, the only open-source compiler for FORTRAN 95 is the GNU project, still yet to be completed. In fact, I understand that there are two seperate projects, one the "original" and one which forked due to personality differences. A Google search for "G95" should bring up one of the two. Both are still in (active) development.

      It is currently possible to get hold of a free FORTRAN compiler; or, to be specific, a compiler for the F language, which is a clean subset of the FORTRAN 95 plus ISO 15581 language standards (the latter bit adds some missing features to FORTRAN 95). F gets rid of much of the FORTRAN 77 baggage in FORTRAN 95 (which exists for backward compatibility), with an aim towards furnishing a good educational language. Nevertheless, it contains all of the "modern" features of FORTRAN 95, and the existence of a free compiler is a bonus.

      --
      Tubal-Cain smokes the white owl.
    3. Re:What language? by sogoodsofarsowhat · · Score: 1

      Use the language best suited to the task at hand. There is no Uber Languge and there never will be. Somethings are just best handled by a specialist.

      --
      . I love the sound of burning women and screaming rubber....
    4. Re:What language? by smitty45 · · Score: 1

      his question (the way I understood it) was what language are the majority of parallel applications written in. some languages lend themselves to being run on these big machines, and FORTRAN is just one of the more popular.

    5. Re:What language? by hawkstone · · Score: 1

      Just a data point for ya: (For reference, I use 3 of the top 5 in the list, though only one is truly in production mode right now).

      In my experience, the majority of the codes running on them are C, but there is still plenty of FORTRAN. But it's typically the newer ones that are C, and older ones that are FORTRAN, and the older ones typically are not using much beyond F77.

    6. Re:What language? by nadaou · · Score: 1

      Another posts describes the incomplete GNU F95 project. http://g95.sourceforge.net/

      You can get the Intel Fortran Compiler, (free for non-commercial use) here:
      http://developer.intel.com/software/product s/compi lers/flin/
      note it works fine on AMD with i686 opts.

      Add'ly, Lahey has a free 30-day trial, and PGI doesn't.

      None of those but the GNU are open-source of course.

      --
      ~.~
      I'm a peripheral visionary.
    7. Re:What language? by AEC216 · · Score: 1

      Yeah, I won't knock Fortran ether, I wrote a Linear Algebra package in (ANSI)C / (GNU)C++ / (1.3)java / (77)Fortran / (?)Pascal. Fortran always rocked the other languages when it came down to speed. Mostly due to the fact that it mapped its memory at compile time and not dynamically. The still teach Fortrash to the mechanical engineering students at the college I went to (Univ. MO-Rolla) and they bitched alot. It is still fun to tell them that the old language they hate so much can crunch numbers so well.

      --
      May I please have my frontal lobotomy if I bring back the ashtrays?
  50. Lucky geek by LiberalApplication · · Score: 1
    I never had the opportunity to be in the presence of anything like that either. The only supercomputers I've ever stood next to have been the discarded husks of a few Crays that devoted geek-acquaintances had scavenged. I always did wonder what it would be like to walk into the space where one of those top 500 machines was kept. What would it sound like? What would it smell like*? ...and would the floor be a-thrum with a deep vibration that made my ears tickle?

    *I just know someone is going to answer, "Mountain Dew, Chee-tos, and poor personal hygeine". Come on? Say it. :)

    1. Re:Lucky geek by Chang · · Score: 1

      I made a pilgrimage to see the Earth Simulator two weeks ago. My aunt lives in Yokohama so it was a good excuse to go.

      They were working on one of the nodes so they let me walk down from the observation window into the machine room for a few minutes. It's a huge room with the the compute nodes in the center and storage all along the outside edges.

      The rooms sounds like high speed rushing water and there is a low vibration from the giant fans underneath that move air through the ventilation system.

      The lighting system is reflective tubes - all of the bulbs are at the end of each tube out of the room so that they don't have guys on ladders changing bulbs above the compute nodes. It makes the room bright enough to see clearly but much dimmer than most data centers I've seen.

      On top of each node there are colored lights that flash during communication events.

      Basically, when you are in that place you know you are in the presence of one of mankinds greatest achievements - especially after you see the visualizations and what they are creating using all that hardware.

      I'm glad I went.

  51. Hmm by Dark+Lord+Seth · · Score: 0, Troll

    An article about supercomputers right after an article about software bloat... Is this a premonition of what we need in 2 years to run Windows Arcadia Pro along with Internet Explorer 7, Office Arcadia Pro and Windows Media Player 11?

  52. Biotech computing cluster by Anonymous Coward · · Score: 0

    I wonder where on the list my companies cluster will show up. IBM is building a 500 node, dual 2GHz PIV Xeon cluster for doing various tasks including DNA research.

  53. "Black Hole University" by Tumbleweed · · Score: 1

    You're lucky you're far enough away from the black hole that is Rolla, Missouri.

  54. i bet by waspleg · · Score: 1

    it's still not playable in software mode (especially for CS/halflife)

  55. FORTRAN, yes indeed. by smitty45 · · Score: 1

    My experience with running applications on these machines (massively parallel) was with the mathematical modelling of dynamic large-deformation phenomena (i.e. car crash simulations) every application I used was written in FORTRAN, and still is. FORTRAN is alive and very well.

    1. Re:FORTRAN, yes indeed. by archen · · Score: 1

      every application I used was written in FORTRAN, and still is

      Now I don't feel that bad not owning one.
      *cringe*

  56. Earth simulator by ehiris · · Score: 1

    I bet the Earth Simulator can't simulate Slashdot earthlings hitting a web site.

    1. Re:Earth simulator by guile*fr · · Score: 1

      i can without even using a super computer, take an egg, an anvil and a very heavy hammer...

  57. I need a sticker... by EverDense · · Score: 4, Funny

    I need a sticker for my PC case that says "My OTHER computer is the Earth Simulator".
    Once I've got that I'll be beating the girls off with a stick.
    As apposed to my current "beating off" activities.

    --
    http://jesus.everdense.com/
    1. Re:I need a sticker... by sean23007 · · Score: 2, Funny

      I sincerely hope the guys in Japan put a sticker on the side of the Earth Simulator that says "My other computer is also the Earth Simulator". Jerks.

      (Hitchhikers Guide)

      --

      Lack of eloquence does not denote lack of intelligence, though they often coincide.
    2. Re:I need a sticker... by Andrewkov · · Score: 2, Funny

      Nah, it probably says "Powered by Honda" followed by a VTEC sticker.

  58. Answer inside by imsabbel · · Score: 1

    No

    --
    HI O WISE PRINCE. WHT TOOK U SO DAM LONG?
  59. WAIT!?! by waspleg · · Score: 1

    what happened to Saddam Hussein's PS/2 Cluster-Of-Mass-Destruction?

  60. Deep thoughts... by gnovos · · Score: 2, Funny

    I wonder what exactly defines a "supercomputer". Wouldn't SETI@Home count? Or in a more abstract sense, all of Slashdot... Except, I suppose, where normal supercomputers are designed to model the destruction of the universe, the "Slashdot machine" is designed to cause it.

    --
    "Your superior intellect is no match for our puny weapons!"
  61. You have never been to japan by gnovos · · Score: 2, Insightful

    For example, you won't see some Japanese Guy driving around in an Escalade with his girlfriend, "just cause".

    Ironically, it's quite a common sight to see Japanese kids driving around in huge American monsters -- with the steering wheel on the wrong side for Japan even! -- "just cause" they think it looks cool.

    --
    "Your superior intellect is no match for our puny weapons!"
    1. Re:You have never been to japan by pao93 · · Score: 2, Insightful

      I agree. I hate it when people speculate about the Japanese young people. Yeah, maybe you see some dorky salaryman driving around in Hybrid. But if it's a young cool guy who likes cars, they'll be driving some monster classic from the US (slightly rare but not uncommon) or a heavily modded, suped up domestic car. Lots of Van type things with wings (?!). Lots of big station wagony cars. Lots of flashing lights. To insinuate that the dudes in these cars have to be Yakuza is fucking pointless. I have a good friend (Japanese) who loves classic american cars and has a huge old 69 Chevy.

  62. Re:They would go back in time by feed_me_cereal · · Score: 1

    What if there were a beowulf cluster of the top 500 supercomputers?

    I suppose, if you truly see a beowulf cluster as a single computer, it would then be the fastest computer and the next 499 fastest would have to be aought out. But then the cluster would have to include these next 499 since it's a beowulf cluster of the 500 fastest. But then... is this a paradox? :) Well, I guess it might be if there were an infinite number of computers, but unfortuneately this recursion has a root :( Actually, I think the answer to your question would be "a beowulf cluster of all computers in the world".

    --
    "Question with boldness even the existence of a god." - Thomas Jefferson
  63. Re:They would go back in time by Anonymous Coward · · Score: 0

    Oh, come on. Everyone knows that God uses a Mac!

  64. I nominate Pink by DeathPenguin · · Score: 4, Interesting

    Pink is the largest LinuxBIOS cluster in the world to date. The only moving parts in each node are cooling fans. This translates into a small savings on hardware, but more importantly means the nodes are by design more reliable thanks to fewer moving parts. Each node is powered by two 2.4GHz Xeon processors with four gigs of RAM and Myrinet 2000 interconnects.

    The raw hardware power, while impressive, isn't what makes this cluster unique. The kicker is in the software, more specifically Clustermatic 3 featuring LinuxBIOS. Stuff happens and nodes fail, but thanks to LinuxBIOS they can be back up in a matter of seconds, not minutes.

    Additional tools for the frontend node from Linux Networx makes updating nodes super-easy. You can flash each node's BIOS with a single command all in a matter of seconds. BProc allows you to run basic shell commands on any node without even installing a distribution on those nodes. w00t!

    What we see here is a big shift away from expensive hardware and proprietary software. The software powering this cluster is 100% GPL, so users save a fortune in software licensing costs alone. And while these P4 nodes in particular aren't exactly cheap, they provide pretty darn good power and are far less expensive than Alpha servers. Also, using the x86 architecture means that consumer boards are not far behind in clustering. In fact, you can check out the LinuxBIOS homepage and see some pretty cheap boards that are supported already. So if you have some spare cash lying around and a couple weekends to kill, you can pick up a cheap board + cpu + memory combo and set it up as a slave node for your desktop machine with the same software these guys use to power this huge cluster.

    1. Re:I nominate Pink by Weirsbaski · · Score: 1

      Pink is the largest LinuxBIOS cluster in the world to date. The only moving parts in each node are cooling fans.

      As opposed to the hampster wheel that's attached to most computers?

      --

      I am not a sig.
    2. Re:I nominate Pink by DeathPenguin · · Score: 2, Insightful

      In many clusters today, each node relies on a hard disk to hold a distribution. This is a real big problem because hard disks generate a lot of heat. This is especially true with 10-15K RPM SCSI drives.

      To add to that, most cluster node manufacturers that I know of design their 1U cases so that the hard disks are mounted on the front of the case for hot-swapping (Which in itself should suggest that people have problems with drives failing). The fans lie somewhere behind the hard disks. For example, this is a picture of an Appro 1122H dual Opteron server. Some cool air (But not a lot) is drawn from the vents on the front-right of the case, but a lot of hot air gets sucked away from the hard disks and blown over the CPUs. A better solution is just to take out the hot swap bays, make sure there's no SCSI backplane in the way, and let the cool air be drawn in from all across the front.

      There are some manufacturers who will put a cardboard heat tunnel around the intake vent, lead it over some blowers, and eventually have to run over the CPUs. That's good design for keeping the CPUs cool, but it basically leaves the hard disks and other components such as north bridge chips and memory modules to fend for themselves when it comes to fighting thermal death. Often times they fail.

      I have yet to see a hard disk that was smart enough to spin down to a lower speed when it reached a certain temperature. Even if they were that smart, who wants an HDD performing poorly when you've already spent ungodly amounts of money for bleeding edge 15K ultra-320 drives? As far as I'm concerned, they're practically built to destroy themselves. This is why diskless cluster nodes are so appealing to me.

      Sorry for my little rant there. I hope it helped to clarify what I meant when I was emphasizing the advantages of having fewer moving parts.

    3. Re:I nominate Pink by joib · · Score: 1

      #3 on the list is a machine made by linux networx for LLNL. It consists of 1100+ dual P4 nodes with quadrics interconnect. And linuxBIOS on all the nodes.

      An interesting thing about this is that it doesn't consist of the ususal 1U rack cases, but they are special cases where the boards are mounted vertically (better airflow).

    4. Re:I nominate Pink by DeathPenguin · · Score: 1

      I certainly like the idea of using effective air cooling like that rather than extreme / expensive methods like dipping the thing in fluorinert.

      The only concern I have is servicing the boards. I've never worked with one of those racks before, but I imagine Linux Networx did a good job at making the boards fairly easy to swap out in case of node failure.

      Here are some pics of the Evolocity 2 nodes in action for everyone to check out.

    5. Re:I nominate Pink by Anonymous Coward · · Score: 0

      Surprisingly some of the top configurations are
      based on old hardware, for example, nearly all
      the most powerful configurations HP claims are
      based on the Alpha processor. It does not
      actually means that the Alpha processor rules
      computing today but at the time of installation
      of the system it surely did. It would be
      interesting to see if clusters based on modern
      technology/ideas like LinuxBIOS would outperform
      the installed ones.

  65. NCSA? by Anonymous Coward · · Score: 0

    Is NCSA (National Center for Supercomputing Applications) still around? I thought they would have made this list, or did they go out of business?

    1. Re:NCSA? by girish · · Score: 1

      31 IBM
      Titan Cluster Itanium 2 1 GHz - Myrinet/ 512 1593.00
      2048.00 NCSA
      USA/2003

  66. But... by Penguinshit · · Score: 1

    if you're counting Deep Thought, shouldn't it be "Terra Flops"?

  67. Albuquerque Journal Article w/ a contrary view. by anzha · · Score: 1

    Emphasis added to highlight some POVs. Note WHO they are too...

    ****

    From the Albuquerque Journal

    Saturday, August 3, 2002

    Sandia May Help U.S. Regain Supercomputer Lead
    By John Fleck
    Journal Staff Writer

    Sandia National Laboratories is negotiating a deal with legendary computer-maker Cray Inc. to build a $90 million supercomputer for nuclear weapons research.

    In years past, this would likely be headlined something like "Sandia tobuild world's fastest computer." But the days of Sandia and the other U.S. nuclear weapons laboratories leading the world in high performance computing are gone.

    The new Cray machine will be tremendously fast, and so big that a $3 million building must be constructed to hold it. But U.S. supercomputers have fallen far behind Japan's best.

    Experts say the Sandia-Cray deal is an important step, however, in pushing the United States back toward preeminence in the supercomputer world.

    The Sandia-Cray deal has the potential to push the state of the art in computer technology, said Horst Simon, director of the National Energy Research Scientific Computing Center in Berkeley, Calif.

    Sandia announced in June that Seattle-based Cray has been chosen to build the labs' new computer.

    Officials will not talk in detail about the deal until negotiations are complete, expected later this month. But in written responses to questions from the Journal, Sandia supercomputer program chief Bob Thomas said the new machine will be at least seven times faster than Sandia's current fastest computer.

    That would make it second fastest among supercomputers currently operating around the world, but still well behind the Earth Simulator, a research machine built by Japanese computer giant NEC for a Yokohama-based research consortium.

    The Earth Simulator shook up the high-performance computing world this spring with a staggering top speed of 36 trillion calculations per second, five times faster than its nearest competitor.

    It was the first time in five years that a U.S. nuclear weapons laboratory has not held bragging rights as "world's fastest." For most of those five years, the title belonged to Sandia.

    "The Japanese now have a trophy on their mantle," said University of Tennessee supercomputer expert Jack Dongarra.

    But more than a trophy, leadership in high-performance computing is important to national security, Dongarra and Simon both argued.

    Sandia and the other two U.S. nuclear weapons laboratories have long pushed the state of the art in high-performance computers in order to run the simulations needed to understand nuclear weapon design, reliability and safety.

    Despite the Japanese leap to the title of "world's fastest," four of the 10 fastest computers in the world are at U.S. nuclear weapons labs, including one at Los Alamos and one at Sandia.

    Among other things, the new Cray machine will be used to simulate warhead performance in the high radiation of a nuclear battlefield, according to Thomas.

    While details about the new computer will not be revealed until the Sandia-Cray negotiations are complete, Simon said it is clear that Sandia is pursuing a different strategy to build the new computer.

    The other large machines being bought by U.S. weapons laboratories are built around clusters of the sort of fast computers used in the business world.

    "The machines we've been buying are nothing more than souped-up webservers," Simon said.

    In contrast, in buying from Cray, Sandia has turned to a manufacturer that builds far more specialized machines.

    After a series of mergers and corporate deals, the Cray of today is very different from the company that in the 1970s built the world's first "supercomputers" at Los Alamos.

    But despite the corporate differences, the move to buy from Cray suggests that Sandia wants a machine designed specifically for scientific applications, Simon said.

    --
    Do you know why the road less traveled by is littered with the bones of the unwary?
  68. that's great, but... by updog · · Score: 1
    Top of the list is the Earth Simulator Center in Yokohama, Japan, with a benchmark performance of 35.86 Tflop/s.

    does it support Ogg?

  69. Correction by istartedi · · Score: 2, Interesting

    The article should read "Top 500 Supercomputes that we know about Ranked".

    Wouldn't you love to know what the NSA uses to crack 128-bit keys? Ever wonder if the solution to RSA-1024 is just laying around in their files somewhere, the employees who know about it sworn to secrecy?

    --
    For all intensive purposes, "whom" is no longer a word. That begs the question, "who cares"?
  70. Tflops/s by bobo+the+hobo · · Score: 1

    Terraflops/second?
    A trillion floating point operations per second per second?
    How upsetting.

    1. Re:Tflops/s by TeknoHog · · Score: 4, Funny
      A trillion floating point operations per second per second?

      It's called hardware acceleration.

      Or perhaps it's the rate at which computers need to speed up constantly in order to run Office.

      --
      Escher was the first MC and Giger invented the HR department.
  71. Slashdotted, as usual ... by rollingcalf · · Score: 1

    They should have used one of the top 500 to host the list of the top 500!

    --
    ---------
    There is inferior bacteria on the interior of your posterior.
  72. elxsi MBOS by lscotte · · Score: 1

    No kidding, Cray always has the coolest looking systems. The Connection Machine was pretty cool too, if I recall, it had 1000's of LEDs in a big matrix showing what was going on inside it, which was mighty purty (I've seen pictures).

    The elxsi MBOS (Message Based OS, amazing I can remember after all this time) was able to run multiple OS's on top of it at the same time. It was the first partitionable system I'd ever heard of. Funny that's coming around again today - both directly (like IBM 690s) or indirectly (like VMWare). Kind of handy to be able to boot a virtual OS to do some testing. Too bad they folded and went to Singapore.

    --
    This post is licensed under the Creative Commons Attribution-Noncommercial-No Derivative Works 3.0 License.
    1. Re:elxsi MBOS by lscotte · · Score: 1

      In fact, here's a picture of the Connection Machine:
      http://mission.base.com/tamiko/cm/cm-image.html

      --
      This post is licensed under the Creative Commons Attribution-Noncommercial-No Derivative Works 3.0 License.
  73. FLOPS are cool... by charlie763 · · Score: 2, Interesting

    Wouldn't it make more sense for processors to be rated based upon their FLOP count? I mean; buying one base upon MHz is like buying a car based upon it's rpm rating.

    And is there any program (preferably linux) out there that will do a benchmark test on my computer in FLOPS?

    --
    Welcome to the land of the free...pay toll ahead...no photography...please open your bag...
  74. Does it qualify by TheDukePatio · · Score: 1

    as a Super Computer if it can't handle being /.'d?

    --
    To Alcohol! The cause of, and solution to, all of life's problems.
  75. Re:They would go back in time by rjoseph · · Score: 2, Interesting

    Uhm...what about me?

    It's number 90 on the new list (was number 85 when it first came out), is entirely self-built by members of the theoretical astrophysics group here at LANL, and (in re: to a comment below) we've even been able to convince LANL to categorize it as a single computer, instead of 294 smaller ones.

    So there you have it, Beowulf in the Top 500.

  76. Where is Apple? by Billly+Gates · · Score: 1
    Steve Jobs said the g4 is a supercomputer, yet I don't see it on the list.

  77. Re:They would go back in time by Q-Hack! · · Score: 1

    çOEãååããã

    Don't you think? It is the gold coin in the cat.

    Or something...

    --
    Some days I get the sinking feeling Orwell was an optimist.
  78. Damn! by gantrep · · Score: 1

    My school's beowulf cluster dropped from 107th to 200th!

    It's still awesome though.

    Prairiefire:
    Specs
    Prairiecam

  79. Yeah, but... by roesti · · Score: 1
    imagine that one of those computers will come on your desktop or something in, say, 15 years. (2^15 = 32768)

    Yeah, but imagine a Beowulf cluster of them...

  80. Re:They would go back in time by feed_me_cereal · · Score: 1

    Um, I'm only a novice Japanese student, and I might have totally mangled this, but let me try to explain as best as I understand. If you know this stuff a lot better than I do, feel free to correct me; I'm finished with my classes and am trying to learn the language on my own now. Anyway:

    This is a Japanese adage which basically means "pearls before swine". The 'ni' in this particular case is, in english, more like "to" than "in"; it has multiple meanings. So, it's more like "gold-coins to a cat". Or it might be more like "gold coins at the location of a cat" which could be more like "gold coins before a cat". Here's the link I read about it at.

    Was that your translation or did you use babelfish? I only ask because it sounds kinda like a babelfish translation. They can be pretty funny :) It's very difficult to translate a lot of Japanese to English (well, for me anyway :). I'm surprised Babelfish does so well but when particles like this are a little ambiguous, it totally fails, which of course is the case at least half the time.

    --
    "Question with boldness even the existence of a god." - Thomas Jefferson
  81. Teraflops? by Anonymous Coward · · Score: 0

    Can we have some decent units as basis for comparison?

    Damn, use bogomips!

  82. Wonder where the F@H network would rank by CracktownHts · · Score: 2, Interesting

    With 80,000 active processors it ought to be up there with the best of them. Too bad it doesn't meet the definition of "supercomputer".

  83. Too bad about Alpha by lophophore · · Score: 1

    HP is winning--with Alpha.

    Too Bad Alpha Is Dead.

    The breakdown by manufacturer statistics would have been a lot different if it was still DEC Alpha.

    Call me a sourpuss.

    --
    there are 3 kinds of people:
    * those who can count
    * those who can't
  84. Where is the SETI network? by Skapare · · Score: 1

    Where is the SETI network? I didn't see it listed. I would think it qualifies as a supercomputer. As SUN has said in the past, "the network is the computer". You can see how many teraflops it averages on the total statistics page.

    --
    now we need to go OSS in diesel cars
  85. Lastest announcement from top500.org by Anonymous Coward · · Score: 0

    They have just stated that 100fps were achieved by only the top 15 supercomputers. The Doom3 bench mark may be kept for sometime, said an annoymous coward from NERSC.

  86. Oops, Forgot one... by paranoidsim · · Score: 1

    0
    G5
    NSA
    32352234626326355860.00
    4235232223626 236626252350960.00
    USA/2003

  87. No SCO? by Z4rd0Z · · Score: 1

    I'm surprised SCO isn't on that list! They are the pinnacle of high end computing, after all.

    --
    You had me at "dicks fuck assholes".
  88. How about FLOPS per Watt ? by staaktdenarbeid · · Score: 1

    Well .. might be a little late now.
    But how would these computers rank in FLOPS (or whatever relevant OPS) per Watt ? How many Operations per Joule ? In the age of portability and embeddedness, I am also interested in getting a maximum amount of computing power in any place I choose.

  89. BOOYAH, I've Got Access to a Top 100 by Salis · · Score: 1

    Oh, it's fast, booyah!

    Computational biology just sucks up those FLOPS!
    It's like budda.

    --
    Favorite /. tagline: "On the eighth day, God created FORTRAN." And it was good.
  90. Most engineering offices could make the list by afidel · · Score: 1

    Near the bottom of the list are a bunch of clusters made with 100 odd workstation class machines on ethernet, so a well wired office of engineering computers could probably make the list. Guess that's why Beowolf is so popular and why US export restrictions are so retarded (not to mention that the Japanese or Europeans will be happy to sell most countries a supercomputer for "oil exploration" or the like) (yes I realize that oil exploration from 3D seismogrophy is a legit use for a supercomputer, it's just that many countries that would want to get their hands on large computers for military purposes would also be able to disguide it as an economic purchase of that type.)

    --
    There are 4 boxes to use in the defense of liberty: soap, ballot, jury, ammo. Use in that order. Starting now.
  91. Where is it? by The+Human+Cow · · Score: 1

    I can't find the HAL9000 anywhere on this list.

    --
    The Human Cow - bringing you scrumtrelescence since 1995
  92. SETI@HOME is actually #2, with 27TFLOPS by billstewart · · Score: 4, Funny
    The second-fastest supercomputer in the world is SETI@HOME, which averages about 27 TFLOPS these days. That's still slower than Japan's Earth Simulator, which is 35 TFLOPS, but it's twice as fast as the Weapons Of Mass Destruction Labs's fastest machine (13.8 TFLOPS), which is about double the speed of Another Weapons of Mass Desctuction Lab's Evil Linux Cluster. So
    1. The World's Fastest Computer is trying to figure out this planet,
    2. The World's Second-Fastest Computer is a volunteer effort to figure out if anybody's on other planets, cure cancer, and do other good things on this planet, and
    3. The Next Fastest Four Computers are trying to figure out how to blow up this planet.
    --

    Bill Stewart
    New Fast-Compression-only CPR http://preview.tinyurl.com/dy575ks
    1. Re:SETI@HOME is actually #2, with 27TFLOPS by SeanAhern · · Score: 1

      As someone work works on some of the computers at LLNL, I'd like to suggest that maybe the simulations that run on these computers have a wider scope than you imagine.

      How about safety calculations? How about high energy physics? How about global climate research? How about new material simulations?

      I think there's a high probability that you may be missing some applications.

    2. Re:SETI@HOME is actually #2, with 27TFLOPS by Noren · · Score: 2, Insightful
      Sorry to interrupt a rant with actual facts, but...

      LANL and LLNL have actually done research on cancer, unlike SETI@HOME which has done no work at all on cancer.

      The University of California is currently a 'Key Sponsor' of SETI@HOME and its Berkeley campus is home to the SETI researchers who set up and use SETI@HOME. The University of California also currently operates both LANL and LLNL.

      I'm not familiar with Evil Linux, is it anything like Red Hat?

  93. Imagine that. by Tokerat · · Score: 1


    Best. Beowulf. Ever.

    --
    CAn'T CompreHend SARcaSm?
  94. Hard to Tell Operating Systems... by billstewart · · Score: 2, Insightful
    The Second Fastest Supercomputer, which isn't on the list, is SETI@HOME, with ~27TFLOPS, and most of the machines it's running on are Windows. Most of the machines don't say what OS they're running, but most of them are running some kind of hacked-up operating system to coordinate communications and tasks. The problem with Windows in this environment isn't whether it sucks, but that it's not open to hacking and customization to anywhere near the extent that most Unix OS relatives are (except apparently SCO, which has trouble running networks on /dev/lawsuit.)

    Also, Windows wants to run on something that looks at least *vaguely* like a PC. Some of these supercomputers look like PCs with odd network peripherals underneath them, some look like clusters of multi-processor shared-memory PCs (sometimes with too many processors for Windows) with a communications layer between clusters. Some of these work ok for Windows (SETI, for instance), while others are too different.

    Also, the communications patterns between nodes and between common applications programs are highly critical here. The tighter the coupling, and the finer-grained the parallelism, the harder it is to fit into whatever framework the operating system provides. Loosely-coupled systems can work just fine on Microsoft; very-tightly-coupled systems need more hacking. And a large part of the Windows plaform is really focused on desktop graphics applications, which simply aren't relevant for supercomputers. (There are people doing clusters with game consoles, such as the Sony Playstation, but that's because they want to use the fast parallel CPU in the graphics engine, not the boring CPU.)

    --

    Bill Stewart
    New Fast-Compression-only CPR http://preview.tinyurl.com/dy575ks
    1. Re:Hard to Tell Operating Systems... by dan_bethe · · Score: 1
      The Second Fastest Supercomputer, which isn't on the list, is SETI@HOME, with ~27TFLOPS, and most of the machines it's running on are Windows.
      That's because the systems running Windows aren't doing anything important most of the time.
  95. NSA's Computers by billstewart · · Score: 2, Interesting
    There's occasional speculation about this sort of thing. For crypto applications themselves, FLOPS don't matter - integer processing and bitmap-bashing do, and it's possible that they've built themselves some efficient DES-crackers or other crypto engines. I doubt that most of those would be Cray-like vector processors - it's more of a job for dataflow architectures and lots of parallelism.

    But the big floating-point applications that NSA has are likely to be signal-processors like Echelon which are trolling for voice signals and such, which is a good match for large numbers of scalar processors. How tightly they're integrated depends on the conveniences of signal collection, which is beyond my ability to speculate credibly :-)

    --

    Bill Stewart
    New Fast-Compression-only CPR http://preview.tinyurl.com/dy575ks
  96. maybe its just me, but clustermatic comments... by seibed · · Score: 1

    maybe its just me, but clustermatic comments will never take the place of Beowulf cluster comments on slashdot.

  97. Deep Thought? no teraflops at all. by astroboy · · Score: 1

    6 * 9 is integer arithmetic.

  98. The correct answer... by Robotech_Master · · Score: 1
    I wonder how many teraflops Deep Thought could have done?


    All of them.
    --
    Editor Emeritus and Senior Writer, TeleRead.org
  99. Re:They would go back in time by Anonymous Coward · · Score: 0

    *shakes head*

  100. Statistical Analysis by Detritus · · Score: 1

    The NSA also does statistical analysis of encrypted data, looking for anomalies that provide valuable clues for cryptanalysts. That can suck up a lot of floating point horsepower.

    --
    Mea navis aericumbens anguillis abundat
  101. ummm.... OFFTOPIC?!?!?!?! by Anonymous Coward · · Score: 0

    Why is this +5 insightful... more like -1 OFFTOPIC... this has nothing to do with the article at all.....

  102. The real question is... by gosand · · Score: 1
    Do they run Linux?

    What? Oh, they do. Well .... woo-hoo.

    No, really, I didn't see any list that showed what OS they run, or at least what OS they are based on.

    --

    My beliefs do not require that you agree with them.

  103. Wait, this can't be right, the Mac isn't on it. by Anonymous Coward · · Score: 0

    I mean, it is the fastest computer in the whole wide world, right? /sarcasm

  104. Luckier geek by LiberalApplication · · Score: 1

    Okay, you're the new lucky geek. That really does sound impressive. It's kind of funny to think that we could have quasi-religious experiences in the presence of machines. I bet I would.

  105. "who works" by SeanAhern · · Score: 1

    I meant "who works". How `bout that "Preview" button, Sean?!

  106. The world's longest ... by Anonymous Coward · · Score: 0

    #51 is split between Indianapolis and Bloomington -- 50 miles apart. Two identical 208-processor Linux clusters.

  107. The real record that HP isn't talking about by mulp · · Score: 1

    HP claims to have 159 of the systems in the top500, but doesn't talk about the fact that over 100 of them are for computer architectures that HP has declared dead, those based on DEC's Alpha and HP's PARISC.

    HP's future is based on commodity chips where the only advantage that HP has is size and the cash to support the significant amount of upfront large scale system integration. And I'm sure that a lot of the folk who have done this work in the past are headed for the chopping block as HP finishes off the parts of DEC that Compaq didn't.