Follow Slashdot stories on Twitter

 


Forgot your password?
Close
typodupeerror
×
IBM Hardware Science

NEC Strikes Back With SX-8 Supercomputer 192

Posted by timothy from the my-honor-student-can-beat-up-your-dad dept.
News for nerds writes "It was just 3 weeks ago that we learned IBM's BlueGene/L with 36.01 TFlops edged out NEC's Earth Simulator, but today NEC announces a new SX-8 supercomputer with a peak processing performance of 65 TFlops (press release). It may be available in the U.S. as Cray's OEM like SX-6."
This discussion has been archived. No new comments can be posted.

NEC Strikes Back With SX-8 Supercomputer More

NEC Strikes Back With SX-8 Supercomputer

Comments Filter:

  • Shallow Minds (Score:5, Funny)

    by stecoop ( 759508 ) on Wednesday October 20, 2004 @01:52PM (#10577529) Journal
    65 trillion calculations per second and all I ponder is if NEC would mind using my user id while running seti@home.

  • Packard Bell (Score:3, Funny)

    by lateralus_1024 ( 583730 ) <mattbaha@gmailLISP.com minus language> on Wednesday October 20, 2004 @01:53PM (#10577536)
    Tom's Hardware stated the the guts are Packard Bell, and it comes with a WinModem. That sucks.

  • Yes, but... (Score:4, Funny)

    by extagboy ( 60672 ) on Wednesday October 20, 2004 @01:54PM (#10577568) Homepage
    ...can it figure out the question to the answer 42?

  • It's like (Score:4, Insightful)

    by Moby Cock ( 771358 ) on Wednesday October 20, 2004 @01:54PM (#10577569) Homepage
    This is like an arms race of yesteryear. The Germans and the Brits with their battleships, the Americans and Soviets with there nukes, the Yankees and Red Sox with payroll. Except this race is way cooler and will likely pay off in a much more productive way.

    • Re:It's like (Score:5, Insightful)

      by Timesprout ( 579035 ) on Wednesday October 20, 2004 @02:08PM (#10577764)
      Except this race is way cooler and will likely pay off in a much more productive way.

      Yes because this sort of computing power holds no attraction to the military for weapon modeling purposes or to the untouchables running echelon type programs.
      • They use DSP to do that. No one knows how to run a AI to do echelon stuff. You could play a big MMORPG on one though. Besides aren't virtual explosions better than real ones?

        • Re:It's like (Score:3, Interesting)

          by Jungle guy ( 567570 )
          The rules to have a job running on the Earth Simulator are very strict. All the results must be made available to the scientific community, under the public domain.

          It leaves almost all research projects with military or commercial purposes out of the Earth Simulator. Most of the data processed there help understand the global weather or the seismic movements. As Japan is a country that has to deal with typhoons and earthquakes, and the Japanese government is ultimately the owner of the Earth Simulator, it

    • And unlike the Yankees and Red Sox there's actual a question over who will win.
    • Who do you thinks buys these systems first?
    • It's interesting to think that the R&D money put into military concerns hasn't had many significant productivity spin-offs. For that matter, the nuclear arms race drove large segments of supercomputer production for years, with the best systems being military bought.
    • And then comes Google with its massively distributed Array of Inexpensive Reduntant Computers (AIRC) and uses all those giga flops to fetch p0rn links and pictures. I'm coming!

      How many flops is 100000 low cost PC:s?

  • Damn.... (Score:5, Funny)

    by identity0 ( 77976 ) on Wednesday October 20, 2004 @01:55PM (#10577589) Journal
    Note that 'SX-8' pronounced 'Sex-ay' :-)
  • It's very impressive and all, but how is this going to benefit me down the line? It's not like they're affordable to small/medium businesses like the Cray [cray.com] or HP's highly valued Alpha DEC workstations [compaq.com].

    We are stiffling progress at the lower level by pricing these systems well beyond the reach of the average researcher or multi-national oil conglomerate.

    Why is this?
    • It's very impressive and all, but how is this going to benefit me down the line? It's not like they're affordable to small/medium businesses like the Cray or HP's highly valued Alpha DEC workstations.
      Are you kidding? A couple years from now you'll be seeing these things, shrink wrapped, on the shelves at Best Buy.
    • That's a dead alpha link (figures). Perhaps you meant this one [hp.com]?

    • Re:Real-world applications? (Score:5, Interesting)

      by Marxist Hacker 42 ( 638312 ) * <seebert42@gmail.com> on Wednesday October 20, 2004 @02:37PM (#10578120) Homepage Journal
      At a rental of only $17,000/month, it's far more affordable than a Cray for small quick projects.
    • Oh, I wish they were, but sadly HP seems to have axed anything with Alpha. I wish Intel would
      take a good look in the mirror and buy the Alpha
      outright (FUD alert: if AMD does it first you are
      history guys...).

      Given the beautiful clean design I wonder if a really die-shrinked version of Alpha would make a nice PDA processor...

      Anyone out there have any clues?
      (sorry, but I just watched Bob Colwell's (ex intel chief architect) lecture to Stanford, so I'm in processor architecture mode right now...

      The answer "L

  • Usage (Score:2, Funny)

    by fembots ( 753724 )
    I was about to say "With a computer this powerful, the world probably only needs 4 of these", but history tells me otherwise :)
    • They expect to sell 700 during its "active lifetime", meaning until sx-9 will be out.
      Not all big versions, but even a 4 node one would have been HIGH in the top500 a few years ago...
    • ...you'd really want more than four if you were going to make a half-decent beowulf cluster of these things.

  • Runs Linux? (Score:4, Interesting)

    by mukund ( 163654 ) on Wednesday October 20, 2004 @01:59PM (#10577648) Homepage

    It seems like it runs Linux as they are claiming that it will use the Global File System [redhat.com] for clustered FS operations.. unless their Global File System is different.

    • Re:Runs Linux? (Score:5, Informative)

      by mukund ( 163654 ) on Wednesday October 20, 2004 @02:07PM (#10577745) Homepage

      In reply to my own post..

      The Cray SX-6 System runs the UNIX-based SUPER-UX operating system. [cray.com]

      Sorry about that. Maybe they ported GFS.. dunno.

      • Re:Runs Linux? (Score:1, Funny)

        by Anonymous Coward
        Just let us know when you want us to join in this conversation, thanks.
      • The GFS fileserver runs on an Itanium Linux server, the TX-7 [nec.com], while the number-crunching computation runs on the SX vector machine.

        Some customers who buy such hybrid SX/TX-7 systems want to use their TX-7 nodes for computation and not just I/O, and so NEC sells software development tools for Itanium systems as well. But the competition in Itanium server space is fierce, with SGI, HP, et.al. having similar offerings.

        • Here is an example GFS configuration [nec.com].

          There is no mystery about "porting" GFS to SX, because it runs on a TX-7 Itanium Linux server, as a dedicated SX peripheral.

          Here is another example [clustervision.com]

          High-speed File Sharing Through GFS: GFS (Global File System) is a file system that enables sharing between several nodes within a SAN (Storage Area Network) environment built on Fibre Channel. By using the TX7 series GFS, NEC offers high-speed file sharing with the SX supercomputer series.

          Imagine a 16-processor TX-7

    • Nope. The Global File System in super-UX is much less general purpose than redhat's GFS. It relies on global shared memory afforded by the full crossbar interconnect of the sx-series computers.

      They are similar from a 1000 foot view, very different in implementation, design, and target user.

  • 65 TFlop is only an estimate (Score:5, Insightful)

    by foobar3149 ( 628047 ) on Wednesday October 20, 2004 @02:00PM (#10577658)
    The 65 TFlop for the SX-8 is only an estimate while the 36 TFlop for BlueGene/L was real performance. So it is not certain that SX-8 will be faster than Blue Gene/L
    • You're trying to say they could have overestimated the performance by almost 100%? Coming from the people that actually built what was until recently the fastest supercomputer, that's extremely doubtful.

      Also, the performance per-CPU and per-node is most likely real data, as they say the SX-8 would ship in December.
      • I just wanted to point out that is not an apples to apples comparison, and that such a comparison is not fair to either machine. For that matter, the number for the IBM machine is only a subset of BlueGene/L so then why not use the 360 TFlop projected performance of BlueGene/L as the comparison point?

      • Re:65 TFlop is only an estimate (Score:5, Informative)

        by Troy Baer ( 1395 ) on Wednesday October 20, 2004 @02:52PM (#10578307) Homepage
        You're trying to say they could have overestimated the performance by almost 100%?

        No, the original poster was saying that the 65TF number from NEC is theoretical peak performance based on the maximum possible number of FP operations per clock cycle (which can never happen in real code, due to pesky little things like having to access memory), while the ~35TF number for the Blue Gene/L prototype is measured performance on an actual piece of code called the Parallel Linpack benchmark. It's not unusual for systems to perform as low as 50% of peak on Parallel Linpack, although 70-90% is more typical on systems with decent memory bandwidth (which the SX8 presumably has).

        (Note that I'm deliberately sidestepping the debate over whether Parallel Linpack bears any resemblence to reality.)

        • Vector systems like SX-8 reach typically 50% of the peak performance in well tuned applications while massively parallel scalar systems crawl at less than 10% (no matter how much man power one invests in tuning). The effective performance of the SX-8 is much higher than that of BlueGene. The peak Linpack performance is meaningless, anyway. HPC Challenge benchmark makes a lot more sense for understanding what performance a machine can reach.

    • Re:65 TFlop is only an estimate (Score:5, Insightful)

      by flaming-opus ( 8186 ) on Wednesday October 20, 2004 @03:51PM (#10578944)
      It's probably a pretty good estimate, as this is just a clock speed bump and packaging update to the sx-6 (earth simulator).

      An equally important criticism is that they've only announced the POSSIBILITY of building a 65TF system. No one has actually ordered one. The cray X1 can scale up to 50TF if fully populated. The X1E scales up to 150TF. This is of no great consequence, as the largest one in production is only 10TF. Yes they could build a really big sx-8, but it cost $200M to build the earth simulator, probably something similar to build this thing.

      There are a lot of computers that are really cool - on paper.
      • Does the NSA still disclose its computing capabilities?
        • Not usually. I've actually done a little bit of work on NSA systems. (you get stack traces by fed-ex after they have been reviewed by security pros) I don't know about ALL of their systems, but the things I've dealt with they are more limited by data throughput, rather than pure number crunching.

          It's a good point that NSA, CIA, etc use big vector boxes, and don't report to top500. They basically bank-rolled the r/d phase of X1 development.
    • The SX8 uses a 90nm CPU clocked two times as high as the one used in the SX6 respective the erarth simulator. Taking into account all the additional improvements it's save to say the SX-8 will be twice as fast as the old one - so 65 TFlops seem reachable.

      For those interested in how vector processing compares with scalar processing in terms of absolute performance and computational efficience, this paper of Leonid Oliker et.al. is definetely worth reading:

      http://www.sc-conference.org/sc2004/schedule/pdfs [sc-conference.org]
      • also, each cpu has 7 times higher io bandwith...
        The 65Tflops estimate is for 512 8 cpu nodes, which less then earth simulators.
        Plus each of the cpus now has 16GFlops (and can archive it with that much memory bandwith it has), but uses less than half of the power of the old sx6.
        So a "earth simulator 2" with 65TFlops would have 20% less cpus and 60% less power consumption then the first.
  • @ Virginia Tech it is rumored the US Gov't is gobbling(harhar) up Apple XServers in hopes to build the "Ultimate" Super computer...

    But then again these are just rumors as Srinidhi Varadarajan is being *highly* recruited
  • IBM has already proved that American technology is, at least, as good as Japanese technology despite all the moans and groans about how we have fallen behind Japan upon the introduction of the Earth Simulator. The good news is that the West (which includes the USA and Japan) have a lock on the most advanced computing technologies.

    Right now, this SX-8 announcement is just a publicity stunt to generate some "shock and awe" in the small supercomputing community of national and commercial research labs.

    Perh

    • Re:NEC SX-8: Predecessor of M-5 (Score:4, Interesting)

      by thesilverbail ( 593897 ) on Wednesday October 20, 2004 @02:12PM (#10577810) Homepage
      The good news is that the West (which includes the USA and Japan) have a lock on the most advanced computing technologies.

      And why is this good news exactly?

      • And why is this good news exactly?

        Obviously you never played through the expansion pack for Battlezone. Let the Chinese get into space and the next thing you know they'll be building cloaked, super-fast hover tanks that kick the crap out of the American units you thought were so powerful in the original game.
    • NEC say in the press release that these super computers start at ~$10,000 a month rental. Maybe that's the base line model but isn't $10,000 a month peanuts for a serious spaceship business?
      • If you have everything prepared, then you get the supercomputer at the end to do your rendering, and in that way, it is cheap.

        Its just idle time.

        Do you have your software written to actually run on it correctly?
        Do the people who analyse the results work for peanuts?
        Have you tested it?
        "Just one more revision" is another 10k.

        Imagine, you get some time on a supercomputer donated to you.

        You sit yourself down at the console, what do you do?

        Well, after writing the obvious pi to 1e999 dp, or for a=0 to 100000
    • Perhaps, the management of NEC should consider generating some "shock and awe" by having Oprah give away free SX-8.


    • Here's your M5 [m5industries.com] for you, right here [m5industries.com].

      Entertainment science at it's finest.

    • IBM has already proved that American technology is, at least, as good as Japanese technology despite all the moans and groans about how we have fallen behind Japan upon the introduction of the Earth Simulator.

      And how many birthdays has the Earth simulator celebrated before IBM finally beat only its Linpack numbers with a low memory per CPU specialized solution? I think BlueGene still loses the HPC Challenge benchmarks [utk.edu]. That isn't what I call winning.


    • the West (which includes the USA and Japan)

      Traditionally, Japan has been considered part of the orient, while the USA has been lumped in the occident.

      I agree that dedicating supercomputer time to investigating new technologies is good.

      Low impact alternative energy sources to supplant existing technologies would be a good start.

    • I suggest that NEC donate computing time on an SX-8 to all the startups designing spaceships (e.g. SpaceShipOne). These startups are short on cash and cannot afford the kind of supercomputer that is needed for modeling the spaceships.

      Did slashdot suffer a timewarp and send the message I'm replying to from 1974 to 2004?

      We have more than sufficient computer power on our desktops to do the maths needed for a designing something like SpaceShip One. What's killing the startups isn't lack of cash, but lack of

  • In other news, scientists create a robot that could replace half of my friends [we-make-mo...ot-art.com].

  • One Liners (Score:1, Funny)

    by Anonymous Coward
    1) I won't go out on a limb but I'll stick my NEC out for this supercomputer.

    2) SX-8? Is that the next in line to SSX Tricky?

    3) As you can see, I am resolute on this war on Tera...and my opponent is clearly a Tera-flopping machine.

    4) Please don't write articles with "$COMPANY Strikes Back" as its title. It begs "Return of the $RIVAL_COMPANY" as a follow-up.

    • Re:One Liners (Score:1, Funny)

      by Anonymous Coward
      I'd mod you up, but there is no "+1, painfully bad jokes."

  • BGL will be back on top in short order (Score:5, Informative)

    by tubbtubb ( 781286 ) * on Wednesday October 20, 2004 @02:12PM (#10577815)

    Remember that the 36.01 TF figure for BlueGene/L was only using 8 racks.
    The final BG/L will use 64 Racks.
    Also, the SX-8 figure is only an estimation.
  • Why?

    Because IBM's Blue Gene/L was/is made with off the shelf parts based on the POWER arch. Meaning it becomes much more efficient dollar wise than the custom made stuff NEC keeps pumping out.

    (on a side note: Where does Lawrence Livermore National Laboratory get all that money to keep buying the latest and greatest super computer ?!)

    Sunny Dubey

    PS: Two of the Blue Gene/L folks will be presenting at my LUG [nylug.org] tonite. I wonder if they will have any responces to this.

    • Re:I'd stick with IBM (Score:4, Insightful)

      by Bishop923 ( 109840 ) on Wednesday October 20, 2004 @02:36PM (#10578117)
      on a side note: Where does Lawrence Livermore National Laboratory get all that money to keep buying the latest and greatest super computer

      LLNL does nuclear research (basically simulating nuclear weapon detonations). We spend $400 Billion on defense per year, what is $200 - $300 million for the latest and greatest super-computer?

      • Re:I'd stick with IBM (Score:4, Interesting)

        by sexylicious ( 679192 ) on Wednesday October 20, 2004 @02:40PM (#10578168)
        LLNL does nuclear research (basically simulating nuclear weapon detonations). We spend $400 Billion on defense per year, what is $200 - $300 million for the latest and greatest super-computer?

        Also, since the US agreed to stop testing nuclear weapons, they've moved on to totally simulating them. Most often the research is done under the guise of astrophysics because the physical processes are almost identical. The US has to stay on top of things, and simulation is the only option to do so.
      • Actually, it's part of the Department of Energy budget, which is $24 Billion. From: http://www.whitehouse.gov/omb/budget/fy2005/energ y .html The 2005 Budget provides $9.0 billion for the national security activities of the National Nuclear Security Administration to include maintaining the safety, security, reliability and effectiveness of the Nation's nuclear weapons stockpile; preventing the spread of materials, information, and technology of weapons of mass destruction by eliminating or securing nu
    • Where does Lawrence Livermore National Laboratory get all that money to keep buying the latest and greatest super computer ?!

      Have you looked at the amount of taxes taken out of your paycheck recently?

  • BlueGene/L is only at 36TFLOPS today, but by next year the full-size version is supposed to clock in at 180TFLOPS. The SX-8 has no chance to survive.
  • which is based on that well proven design of ropes and pulleys in Apraphulia (see Sci Am a long time back), I wonder if they will have heat dissipation problems from the friction...
  • Even with 65Tflops it still takes 3 min to apply my desktop settings.
  • Has it come down to this? A TFlops-size contest?

  • But.. (Score:5, Funny)

    by SnarfQuest ( 469614 ) on Wednesday October 20, 2004 @03:04PM (#10578465)
    How well does it handle XP SP2?
    • On a related note, it is the first computer reported to be able to play DOOM3 at a playable framerate.
    • Better than it would handle Longhorn. But we'll probably have a petaflop machine by the time that's released, so we should be fine.
  • Comment removed based on user account deletion
  • WooHoo! Just in time for Halo2/Half Life 2!!!
  • Two IBM's BlueGene/L.
  • Cool,
    Now ID soft can release Doom3 written in VB! I bet this thing willl get close to 30 fps!

Slashdot Top Deals

Our OS who art in CPU, UNIX be thy name. Thy programs run, thy syscalls done, In kernel as it is in user!

Close