Slashdot is powered by your submissions, so send in your scoop

 


Forgot your password?
Close
typodupeerror
×
Intel Hardware

Teraflop In A Box At SC2003 114

Posted by timothy from the worth-it-for-the-slideshow dept.
HPC Prophet writes "For those of you that can't go to SC2003 or can't afford the US$750 late registration, here is a small taste of what we put together for our friends at Mellanox Technologies...It benches out at over 1.2TFLOP (192 dual Intel Xeon Processor blades, 64 in a Rackable chassis, 128 crammed into a Ciara chassis and all connected via InfiniBand) and loaded up with Callident Rx (based on NPACI Rocks) OS/Middleware. Total estimated time to unpack, build and get up and running was 17 hours." Read on for some details on this power-hog.

"We had the single-most power density for the smallest size booth they offer (380amps @ 208v in a 5U of rack space (look closely at the bottom of the middle rack containing all the cables and InfiniBand switches). Cooling was very nice too, we maxed out our Liebert HVAC when building it initially. Oh, by the way, this would end up somewhere in the neighborhood of #38 on the June 2003 Top500 list. There are a couple of other pictures on there too of some of the other attractions at SC2003 like the 128-node cluster that NPACI folks will build in a 2 hour period. Sorry about the cheezy slide show, I had to be quick."
This discussion has been archived. No new comments can be posted.

Teraflop In A Box At SC2003 More

Teraflop In A Box At SC2003

Comments Filter:

  • Awsome (Score:4, Funny)

    by nberardi ( 199555 ) * on Wednesday November 19, 2003 @08:48AM (#7509902) Homepage
    I like that they actually put this demo together with Windows XP Power Toys.
    • --I'm definitely not an MS fan, but hey - it was functional.

      --What I'd like to see is an article on the *software* they used. Saw some interesting screenshots, would like more depth.

      --I've always been kinda curious about clusters but:
      o What can you *do* with them besides graphics?
      o What's more important - Memory, HD speed, or Proc speed?

      --I'd like to compile my Linux kernel on a cluster and see how much faster it is, but have no idea how to go about doing it. I have 3 machines - P233MMX, AMD900Duron, a

  • For those ... (Score:4, Funny)

    by Rico_za ( 702279 ) on Wednesday November 19, 2003 @08:55AM (#7509926)
    "For those of you that can't go to SC2003 or can't afford the US$750 late registration"
    What about those of us that don't have a clue what sc2003 is?
    • High performance computing, networking and storage conference

    • Re:For those ... (Score:3, Informative)

      by deanj ( 519759 )
      SC2003 is Supercomputing 2003. They hold this conference every year around this time.

      Unless you're REALLY into supercomputing (and these days, it's mostly cluster stuff), this isn't exactly the most exciting conference you can go to.
    • "What about those of us that don't have a clue what sc2003 is?"

      If you have a computer geek membership card, turn it in. If not, proceed directly to the next article. Do not pass go, and do not collect 200 miscellaneous promotional trinkets.

      Although late registraton for exhibits is only $80, and it's $700 for the tech program.

  • LINK for SC2003 (Score:5, Informative)

    by Danathar ( 267989 ) on Wednesday November 19, 2003 @08:59AM (#7509936) Journal


    In case anybody wants it, the link to the conference is at

    http://www.sc-conference.org/sc2003/ [sc-conference.org]

    Several of the lectures are being broadcast via high bandwidth video if
    you are on Internet2.

  • Yet Another Unremarkable Cluster (Score:3, Insightful)

    by Anonymous Coward on Wednesday November 19, 2003 @09:00AM (#7509941)
    A box full of Pentium Xeons in a cluster. So what? This stuff is getting rather passe. Where is the invention and innovation?

    • Re:Yet Another Unremarkable Cluster (Score:5, Insightful)

      by Zocalo ( 252965 ) on Wednesday November 19, 2003 @09:05AM (#7509959) Homepage
      Where is the invention and innovation?

      Perhaps you missed the bit about building the world's 38th most powerful computer (based on June '03 figures) in 17 hours? Damn impressive by any counts.

      • Where is the invention and innovation?

        Perhaps you missed the bit about building the world's 38th most powerful computer (based on June '03 figures) in 17 hours? Damn impressive by any counts.

        Well, since June '03 approximately 3600 hours have passed; 211 times 17 hours - who knows what the competition is up to ;-)

      • Re:Yet Another Unremarkable Cluster (Score:1, Informative)

        by Anonymous Coward
        First off this system is NOT the 38th most powerful computer; in fact as far as I can tell they (Rackable) aren't even on the current top500.org list. Oh, too bad couldn't get linpack to run?

        Glenn Oterro is a devious bastard that likes to call himself "{foo} Prophet"; his email sig claims Linux Prophet, on /. it is obvious he is the so-called HPC Prophet. Take your head out of your ass buddy.

        He is an egotistical ass-clown that likes to promote his yet-another-useless-HPC-poser-company (Callident) to the
      • Does it do any useful WORK?

  • Why Xeon? (Score:1, Interesting)

    by Anonymous Coward
    I though itanic was supposed to be wonderful according to intel and HP. So why are they not promoting huge clusters of itanics? Why are they talking terraflops with cheap and nasty Xeons? 32-bit Xeons?! Everyone else is 64-bit nowadays.

  • Damnit! (Score:2, Funny)

    by soliaus ( 626912 )
    So THEY are the ones who stole my InfiniBand fiber christmas light display!

    Rotten kids, cant trust 'em these days.

  • http://www.testdrivehpc.com/sc03/SC2003_booth_1011 _TFLOP_Cluster/html/35.htm

  • At long last (Score:3, Funny)

    by Anonymous Coward on Wednesday November 19, 2003 @09:06AM (#7509969)
    a computer that will be able to run Windows Longhorn!

  • More like #84 (Score:5, Interesting)

    by Hew ( 31074 ) on Wednesday November 19, 2003 @09:19AM (#7510020) Homepage
    If you look at the more recent November 2003 list [top500.org] instead of the older June 2003 [top500.org] one, this cluster would rate more like #84 than #38.
  • My girlfriend's rackable, but she doesnt clock anywhere near 1.2 taraflops. More like a few hertz, but hey...I can dream cant I?

  • Site slowing considerably... (Score:3, Funny)

    by Phil John ( 576633 ) <phil@webstars[ ].com ['ltd' in gap]> on Wednesday November 19, 2003 @09:52AM (#7510172)
    ...something tells me that they aren't running it on their 1 tflop box. ;o)
  • Sure would be nice to update more then one document, write and deploy some code and read email with out getting a blue screen of death.

    But then I am SURE windows would bring this box to its knees

  • How well do these blade boxes stand up to full trottle usage? Would a box like this handle running the distributed.net client [distributed.net] for days and weeks and years? Although because this is an Intel [distributed.net] box they will be slow as compared to AMD [distributed.net], but still a valid question.

    • 1. There are much more important things to do with a computer. And there are very little things less usefull than distributed.net...
      2. What do you really think is the point to create a >100nodes cluster? A tip: It's not running idle.

      • Blasphemy!

        My point was this. Can a blade setup like that take a 100% load for extended periods of time? Also, if my 100 node cluster is not running apps all the time, why not run ./dnetc on it to eat up the wasted cycles?

        • Its not a server rack for webhosting, its a HPC system. Anything else than 24/7 at full load and it would be able to do its job.

          btw: it doesnt matter if you have all systems running at full load for a day or a year, after 15 minutes the rack has reached full temperature, and if its not to hot then, it wont be a few weeks later...
  • http://www.sun.com/2003-1118/feature/

  • I thought the title read SCO2003.

    Then I laughed out loud at the absurdity. SCO doesn't make products.

  • why oh why, what happened to the news these days, seriously this just seems like one big advert and it is happening more and more at the moment.

    I know that i will get trolled for this but i wanna read kewl stuff, not about #shock# a fast server (thats not even that fast really)

    oh well mod me down i can afford it (as long as my karma repayments are ok )

  • IBM: more TFlops, smaller box (Score:5, Informative)

    by RalphBNumbers ( 655475 ) on Wednesday November 19, 2003 @11:14AM (#7510764)
    Meanwhile, IBM recently built the prototype for a single BlueGene/L node, and it manages to cram 1024 PPC440 processors, with a Rpeak of 2Teraflops, and an Rmax of over 1.4TF into about half the space of the full racks mentioned in this article.

    While this article is obviously about a somewhat less custom system than BlueGene/L, I'd have to say I'm much more impressed with IBM's achievement.
    • True, but the BlueGene/L node is not stackable you can't put one on the top of another, so in real world cases these systems will use close to the same amount of space.

      And this system cost ca. $1 mill. while I guestimate that a BlueGene/L node will cost $2-4 mill.

  • All these low powered clusters are fine and well, but what is the state of supercomputing for problems that really arent parallizable?
  • I have to say after reading up on the "Rocks" cluster OS software they were using, aside from extravigant benchmarks, and bragging rights most of these multi-node cluster "supercomputers" are fluff when it comes to the average users's applications.

    I buy a dual CPU or A Quad CPU machine for example because I know when I run a multithreaded app in XP or 2k or linux it'll spread out the load on all the cpus.

    Just about all of these cluster programs are a complete pain in the ass, and either required specially
  • But back at SC96, I remember paying a nice cheap $75 to get in the door. Quite a bit of inflation, there.
  • Until the box that teraflop comes in is no larger than a desktop full-tower case and costs under $5,000, it's all somewhat esoteric for the average Joe.

    As for what all that power is good for... Why do you need a use in advance of the power? Do you think there was some proto Les Paul sitting around in the 1700s with a solid body guitar and pick-ups, thinking "if someone would just discover electricity, this baby would wail"?

    Make the power available and people will literally hurt themselves coming up wi

Slashdot Top Deals

fortune: cpu time/usefulness ratio too high -- core dumped.

Close