Follow Slashdot stories on Twitter

 


Forgot your password?
Close
typodupeerror
×
Supercomputing IBM Hardware

Largest Privately Owned Supercomputer 292

Posted by timothy from the beats-the-basement-beowulf-rack dept.
GORMUR writes "IBM has launched its Watson Blue Gene system, the largest privately owned supercompuer seen by the press. The super computer is described reaching a whopping 91.29 teraflops. IBM has plans on giving Academic researchers access to some computing time. Some more info can be found the IBM site. All this makes you wonder what other supercomputers are out there, not known to the press, and if it's time to increase the size of your private key and strengthen your encryption."
This discussion has been archived. No new comments can be posted.

Largest Privately Owned Supercomputer More

Largest Privately Owned Supercomputer

Comments Filter:

  • REVENGE! (Score:5, Funny)

    by flawedgeek ( 833708 ) <karldnorman@nospAM.gmail.com> on Tuesday June 14, 2005 @08:03PM (#12819552)
    Apple, you might wanna rethink that switch to Intel.....

    • Re:REVENGE! (Score:5, Funny)

      by ottothecow ( 600101 ) on Tuesday June 14, 2005 @08:16PM (#12819650) Homepage
      And you know that that offer to academic institutions is just a lure...

      "You see, this is the finest private super computer on IBM processors. You may have heard about the school that has a apple super computer, well, that was made when they used IBM processors. If you happen to be in the market for a supercomputer be aware that you can no longer trust Apple to make them the say way and are safer going directly with a system from IBM. Please enjoy your computing time, we sincerely hope that you did not accidentally underestimate the power of our supercomputer and lease too much time."

      All inquiries can be addressed by the sales division in room 341b.

    • Whoah! (Score:2, Funny)

      by minginqunt ( 225413 )
      That's like as powerful as *fifty* PlayStation 3s, all working together!!!

      Can you imagine?

  • NSA... (Score:5, Interesting)

    by ThomasFlip ( 669988 ) on Tuesday June 14, 2005 @08:08PM (#12819575)
    I think it's safe to say that the NSA, with it's largest budget out of any intelligence agency in the U.S, has probably cracked the 100 TF mark ? It's a shame we will never no what kind of muscle they can flex.

    • Re:NSA... (Score:5, Informative)

      by slavemowgli ( 585321 ) on Tuesday June 14, 2005 @08:16PM (#12819646) Homepage
      What's really bad (or good, depending on one's point of view, I assume) about the NSA is not just the computing power they likely wield (they're the biggest consumer of electric power in the entire state of Maryland, so they probably do have some big iron there on site), but also the theoretical power in the form of mathematicians. The NSA is the single biggest employer of mathematicians in the world, and it's probably safe to say that they are at least a couple of years ahead of the rest of the world as far as cryptography and cryptanalysis is concerned.

      Remember, for example, that the NSA invented public-key cryptography before Diffie and Hellman did; or remember that they made some changes to the S-boxes for DES when it first was submitted that noone understood back then but that did turn out to eliminate weaknesses in the original design later on.

      I dare say that this theoretical advantage is actually more important than the pure number crunching power they wield. There's virtually no limit on the computing power you can buy if you have enough money at your disposal (for example, there is no real reason why IBM shouldn't be able to build a system roughly a thousand times as fast as the BG/W system if someone paid the necessary 40 billion dollars), but you can't buy advances in mathematics with money.

      • What you can't buy with money (Score:5, Interesting)

        by antispam_ben ( 591349 ) on Tuesday June 14, 2005 @08:40PM (#12819815) Journal
        The NSA is the single biggest employer of mathematicians in the world, and it's probably safe to say that they are at least a couple of years ahead of the rest of the world as far as cryptography and cryptanalysis is concerned. ... but you can't buy advances in mathematics with money.

        Then what do they use to pay their mathematicians? Coffee?
        • That's beside the point. The point *is* that you can't just go out and buy knowledge and scientific progress like you can buy fast computers.
          • you can't just go out and buy knowledge and scientific progress like you can buy fast computers.

            Sure you can. It's called funding the research. the more you fund, the more likely it is that you end up backing a real winner. But if the mathematician has to teach at a high school or drive a taxi to feed himself, well, there won't be much progress made now will there?
            • Well well well. It seems that there really is some intuition in the idea that mathematics cannot be purchased per unit like fruit in a grocery store.

              Paying for mathematic research is definitely liable to result in quality mathematical ideas, but intuition about the intangibles:

              A. We may hit a barrier that no real gain in mathematics can be obtained regardless of how much money is humanly available.

              For instance, suppose computers become intelligent and perform mathematic discovery very quickly. In the lim
        • but they can buy them a 120 ft Cabin Cruiser that they can use to pull up along side of it.

        • Then what do they use to pay their mathematicians? Coffee?

          Yes, in a way. I have a friend who is just INTO that sort of thing and wants nothing more than to have a fat paycheck for just being some guy who can figure that stuff out in his head.

          So far, he's headed in that direction, he does super-low-level math for his university and the NSA for free under his professor, and he enjoys it.

          Good coffee can go a long way if there's a reasonable expectation of a similar paycheck behind it.

          And BTW, John, keep
        • I do want to point out that if you are a mathematician with the ability to seriously advance mathematics, it would totally suck to not be able to publish any of your major results. If a high stature mathematician is willing to work for NSA and risk not being able to publish work which he/she has done in his capcity as a NSA researcher, he/she most likely is in it for more than just money.

          I do wonder, suppose some NSA guy proves the Riemann's hypothesis. What would they do? How far does patriotism go?

      • Re:NSA... (Score:2, Offtopic)

        by imr ( 106517 )
        ...

        The NSA is the single biggest employer of mathematicians in the world

        ...

        but you can't buy advances in mathematics with money.
      • remember that they made some changes to the S-boxes for DES when it first was submitted that noone understood back then but that did turn out to eliminate weaknesses in the original design later on.

        you're probably right about the nsa being years ahead of the rest of the world, however the above isn't a good example of that because that little episode (des becoming public) pretty much started public cryptographic research.

      • ...the NSA has a "mission impossible". They are supposed to a) Secure national systems with unbreakable cryptography and b) Be able to break cryptosystems. So how do you provide one, without really giving up the ability to the other? You can try operating in that really narrow space between "what a layman/corporation/hostile government can do" and "what NSA can do" but well, that is a pretty slim corridor.

        Being able to throw 1000x the computing power on it is a dangerous game to play - that would be "just
    • No Such Agency
    • ... that NSA would be interested in teraFLops? Last time I checked, their kind of processing required manipulating bits (in weird ways), not imprecise floating point numbers. Go to DOE to pay for FLOPS...

      Not that I'd know, but I can still guess... ;-)

      Paul B.
    • heh, I could think of a test of their abilities that involves using less & less complex encryption until the nice g men kicks in your front door....then you'd have an estimation of their computing power.

  • SHA Collisions (Score:2, Interesting)

    by Anonymous Coward
    People using 256-bit encryption algorithms should be safe for now, given the massive amount of computations needed for key exhaustion. However we should be working on implementing SHA-512 as soon as possible as it will soon become trivial to find collisions in SHA1

  • My data isn't really all that private (Score:3, Insightful)

    by Lingur ( 881943 ) on Tuesday June 14, 2005 @08:08PM (#12819579)
    If you have to defend yourself against some entity that owns the world's fastest supercomputer and doesn't want you to know it, I don't know what you'r e hiding and I don't want to know.

    Seriously, I'm not about to change all my passwords and strengthen my keys because whatever money I have in my bank account is just a drop in the ocean for those guys.

  • A 'what' 91.29 teraflops? (Score:5, Funny)

    by metachor ( 634304 ) on Tuesday June 14, 2005 @08:09PM (#12819586)
    Is 'whopping' really the only adjective adequate enough to describe supercomputer performance?

    Google search of 'supercomputer whopping [google.com]'.

  • Pish posh... (Score:5, Funny)

    by AtariAmarok ( 451306 ) on Tuesday June 14, 2005 @08:09PM (#12819591)
    Compared to the Milliard Gargantubrain in my garage, this thing is a mere abacus. Consider it not.
    • I was thinking sopmethign simular. My pieced together 200 processor unit should reach a couple dozen terraflops. I know it can display 9 pron dvds and 2 downloaded pron videos all at once. the only thing it lacks is a spell cheker.

      Maybe i too can rent proccessor time out to universities in exchange for a faster internet conection. Just imagine, pron faster then i can watch. heaven can be real.

  • I have a supercomputer (Score:5, Funny)

    by Nicky G ( 859089 ) on Tuesday June 14, 2005 @08:10PM (#12819593)
    In my pants.

  • So... (Score:4, Funny)

    by codexwriter ( 682385 ) on Tuesday June 14, 2005 @08:10PM (#12819597) Homepage Journal
    I need 400 PS3's to make one of these.
    Who wants to help me start a fundraiser?

  • "Seen by the Press" (Score:5, Funny)

    by DyslexicLegume ( 875291 ) on Tuesday June 14, 2005 @08:11PM (#12819605)
    Notice how the article says "seen by the press"...maybe there's an even more powerful one in the hands of some evil mastermind on an island in the Pacific who is plotting world domination by having it create a super weapon to destroy everything in its path...yet the computer always keeps giving the same answer:

    42
  • What's the largest non-privately owned supercomputer? And can I play Doom 3 on it?

  • sure. (Score:5, Insightful)

    by St. Arbirix ( 218306 ) <matthew.townsendNO@SPAMgmail.com> on Tuesday June 14, 2005 @08:12PM (#12819612) Homepage Journal
    All this makes you wonder what other supercomputers are out there, not known to the press, and if it's time to increase the size of your private key and strengthen your encryption.

    Yes, I, private citizen of a nation with a resident population of 296,365,988, am worried that the stuff I use private key encryption on will be under attack.

    Until I'm dating a girl with a billionaire ex-boyfriend/stalker I think I should be fine keeping things the way they are.

    Besides, I tend to make up my own encryption scheme for truly sensitive pictu^H^H^H^H^Hdocuments and then just delete the method.

    • Re:sure. (Score:4, Informative)

      by Anonymous Coward on Tuesday June 14, 2005 @08:49PM (#12819869)
      Encryption Schmichion

      We should be more concerned about someone with a Knoppix live-cd or something along those line. That thing has thus far given me unfettered access to more than 98% of the computer system into which I have booted it. For those that had no CD-rom you can just use an USB cd-rom. The only trouble you might encounter is if the BIOS is not set to boot off the CD-rom first and it has a password

      • Re:sure. (Score:2, Informative)

        by haakoneide ( 788114 )
        Not a problem at all. Use http://btmgr.sourceforge.net/ [sourceforge.net] to boot from the CD-rom using a floppydisk
      • We should be more concerned about someone with a Knoppix live-cd or something along those line. That thing has thus far given me unfettered access to more than 98% of the computer system into which I have booted it.

        If you have physical access to a computer, it's hardly surprising that you can gain full control over it. Similarly, if you give someone else physical access to your computer, it's not your computer anymore.

        Of course noone should be storing sensitive data in plaintext on a workstation's hard d

    • Besides, I tend to make up my own encryption scheme for truly sensitive pictu^H^H^H^H^Hdocuments and then just delete the method.

      Dude, if you encrypt your pr0n with a proprietary, undocumented encryption, you're either wearing enough tin-foil to build an aircraft, or you're into some really wierd shit I don't wanna know about.
      • You missed the girlfriend comment in the previous paragraph.
    • "Yes, I, private citizen of a nation with a resident population of 296,365,988, am worried that the stuff I use private key encryption on will be under attack."

      The more processing power you have, the more insignifigant the effort to crack a single encrypted message. "Security through obscurity" doesn't help against brute force attacks.

      Whether or not someone will actually be there to interpret the data is another matter, but why let it get to that stage?


    • "Until I'm dating a girl with a billionaire ex-boyfriend/stalker I think I should be fine keeping things the way they are."

      No matter how teraflop-sophisticated the tools of humanity become, it always comes down to "pussy".

  • Dearest Slashdot Readers, (Score:4, Funny)

    by Anonymous Coward on Tuesday June 14, 2005 @08:14PM (#12819628)
    It is largely unnecessary to increase the size of your keys, it stopped slowing us down quite a while ago. Don't even get us started on the usefulness of tin foil hats.

    Love,

    The Government

    P.S: Don't you people starting clearing the porn off your hard drives, this job gets pretty boring sometimes.
    • Now, come on, Mr. G-man: we know love is something you've never Felt...
    • Long ago and far away in a distant galaxy, when I was in high school, there was an IBM S360 mainframe that was connected to several 3270 terminals in the library. I had typed in quite a collection of obscene limericks at one point and was in the middle of adding some new ones when the multitasker decided to crash. Apparently when this happened all 60 kwords of magnetic core were automatically dumped to the computer room line printer. Normally this system would come back up right away, but it didn't for the
  • Out of curiousity, how much is the average supercomputer utilized? I mean, out of a 24 hour day, how much of that time does the supercomputer stay at >50% utilization? What is considered "full" utilization? Every CPU at >x% load? y of z CPUs at >x% load?
    • I would think (given the way many of them lease out time) that many supercomputers run at pretty high utilization most of the time.

      People prepare the simulations or calculations that need to be run and estimate a block of time needed. They then give that to the operators and request said block of time. The operators run the simulation as soon as that block is available.

    • Re:Anyone out there care to comment? (Score:5, Insightful)

      by saratchandra ( 847748 ) on Tuesday June 14, 2005 @08:24PM (#12819698) Homepage
      I've been using several supercomputers for my research project. Most of them are very busy. Eg. On the IBM P690(Cheetah) at Oakridge National labs,you have to wait for a week to get your 512 processor job scheduled. This is an extremely busy system. On the other hand,you have systems like the Itaniun cluster at NCSA(National Center for Supercomputing Applications) which schedules your jobs a lot quicker. Actually you can check out the usage of this cluster online at http://tg-monitor.ncsa.teragrid.org/ [teragrid.org] (don't slashdot it, it is quite useful to a lot of researchers :-) )
    • My guess is that most super computers are at 100% load. The reason being is that after you have spent millions on a new computer system that will be obsolete in two years, you want to get all the utilization out of it you can get in that time. Of course, this probably requires careful planning to queue simulations into the machine to reduce idle time.

      It's not like there is one screen with a researcher typing his code away at it. They probably test their code on lesser computers (a simple array of desktops)

  • I don't like SPAM (Score:5, Funny)

    by ankhcraft ( 811009 ) on Tuesday June 14, 2005 @08:18PM (#12819662) Homepage

    All this makes you wonder what other supercomputers are out there, not known to the press, and if it's time to increase the size of your private key and strengthen your encryption.

    Increase the size of my private ... and strengthen ... wait a sec! Ya' trying to sneak some SPAM past us?

  • As impressive as it is (Score:4, Funny)

    by hobotron ( 891379 ) on Tuesday June 14, 2005 @08:22PM (#12819687)

    Im waiting for Sherlock Holmes Blue Gene system.
  • I have a complete, operational, DG MV30000 [simulogics.com]. And for private ownership by a civilian, that's pretty cool..

    (the link is to a sales brochure page)
  • I hope IBM didn't have to pay a lot for that. I mean that's what, only 45 Playstation 3's?

  • Nice! (Score:3, Interesting)

    by Duncan3 ( 10537 ) on Tuesday June 14, 2005 @08:47PM (#12819859) Homepage
    Almost 1/2 a Folding@home, I'm 1/2 impressed ;)

    Holy interconnect batman!
  • I only need another 2,591,501 of you to sign up for just a measly, diminutive, insignificant, minuscule, teensy-weensy little 3,520,725 offers so I can get one of these whopping supercomputers for free!

  • DC? (Score:2, Interesting)

    by ilyanep ( 823855 )
    How come nobody counts Distributed Computing as Supercomputers? I'm sure many of the BOINC Projects (SETI@Home at berkeley, E@H at UWM, etc.) have close or even higher than that.

    • Re:DC? (Score:2, Insightful)

      by Kontinuum ( 866086 )
      I suppose part of it is the difficulty in benchmarking. It's a hard enough task when you've got all your processors sitting in front of you. With a distributed computing system, you can't very well ask everyone to not touch their computer for a few hours on Wednesday at noon. Additionally, it takes a long time to distribute information across distribution computing systems. So your timing would be subject to all the events that affect network speed around the world. In the end, you'd probably find that

      • Re:DC? (Score:3, Interesting)

        by mvdw ( 613057 )
        Perhaps a decent measure would be to average the processing out over a week or so; eg each seti@home unit is, say, 1e9 floating point operations, calculate how many units are processed in a week, divide by seconds in a week, there's your number. This method allows for the redundancy of the @home method, ie, each unit will be computed a number of times, if only the completed units are counted it gives a measure of true (sustained) performance.
      • I would suggest some standardised physics application, perhaps a model of a set of different shapes of windmills, at a set of wind speeds, driving a water pump.

  • My Computer Running Windows (Score:3, Funny)

    by kai.chan ( 795863 ) on Tuesday June 14, 2005 @10:01PM (#12820319)
    Fastest privately owned supercomputer? That would be my computer running Windows. It has a record of Always-Flops.
    • Fastest privately owned supercomputer? That would be my computer running Windows. It has a record of Always-Flops.

      Sounds like your computer needs some of that "Virtual Viagra" I keep getting e-mails about.

  • I've been trying! (Score:3, Funny)

    by ScrewMaster ( 602015 ) on Tuesday June 14, 2005 @10:33PM (#12820481)
    ... if it's time to increase the size of your private key ...

    I've been trying to increase the size of my private key, but those little blue "enhancement" pills didn't do anything for me.

  • Meanwhile (Score:2, Offtopic)

    by Orion Blastar ( 457579 )
    IBM couldn't market themselves out of a paper bag since they used Charlie Chaplin to represent their PC line of computers. No wonder Microsoft took a lot of the market away from them.

    Will this become the world's fastest privetly pwn3d Supercomputer once it is on the Internet? Got Unix exploits and script-kiddies?
  • The largest privately owned supercomputer is the one in my secret underground laboratory.
    Oops! ^H^H^H^H^H^H^H^H^H^H^H^H^H^H^H^H^H^H^H^H^H^H^H^H^H ^H^H^H^H
  • All this makes you wonder what other supercomputers are out there, not known to the press, and if it's time to increase the size of your private key and strengthen your encryption.

    Real Security starts when you don't use computers to transfer the data. Very sensitive data (to the holder) frequently goes by encrypted, time limited, self destruct if handled wrong, media paths. Not over the Internet. Sneaker net on steriods, in other words.

    What kind of data is handled that way? One time pad transfers for ba

  • I want to know what happens to old super computers. Then can be used for quite (10 years max?) while since the parallelism is what's more important than the individual raw processor speed, Do they sell them or what?

    • Re:Old Supercomputers (Score:4, Interesting)

      by joib ( 70841 ) on Wednesday June 15, 2005 @01:38AM (#12821148)
      Well, the Cray T3E they used to have at the supercomputer center where I submit my stuff was dismantled and the pieces thrown into a big trash bin in the yard. *sniff*

      The life of a supercomputer is AFAIK really closer to 5 years than 10. It's not that they aren't impressive machines even 5 years old, it's just that they use _lots_ of power and floor space. Looking at how much computing per $ you can do, it's just cheaper to replace them with something new than to keep them running.
  • nah, can't say it.

Slashdot Top Deals

Single tasking: Just Say No.

Close