Become a fan of Slashdot on Facebook

 



Forgot your password?
typodupeerror
Sun Microsystems Supercomputing

Sun Super Computer May Hit 2 Petaflops 134

Posted by ScuttleMonkey
from the coming-out-swinging dept.
Fletcher writes to tell us that Sun Microsystems has revealed their "Constellation System", a new supercomputing platform that the company hopes will put them back in the running for top dog in the supercomputer race. "The linchpin in the system is the switch, the piece of hardware that conducts traffic between the servers, memory and data storage. Code-named Magnum, the switch comes with 3,456 ports, a larger-than-normal number that frees up data pathways inside these powerful computers. 'We are looking at a factor-of-three improvement over the current best system at an equal number of nodes," said Andy Bechtolsheim, chief architect and senior vice president of the systems group at Sun. "We have been somewhat absent in the supercomputer market in the last few years.'"
This discussion has been archived. No new comments can be posted.

Sun Super Computer May Hit 2 Petaflops

Comments Filter:
  • Good to see! (Score:2, Interesting)

    by Anonymous Coward
    Great job Sun!

    Once again you have shown us the power of talent, determination, and skill.

    Rock Rock On!
  • Obligatory (Score:2, Funny)

    by CompMD (522020)
    ...but will it run linux?
  • Zoolander (Score:4, Funny)

    by Anonymous Coward on Wednesday June 27, 2007 @04:18PM (#19668549)
    I dearly, dearly hope that the followups to Magnum are codenamed LaTigra and Blue Steel :)
    • I doubt Sun would come out with a system named Blue Steel. IBM, on the other hand, may be looking for something to call whatever they come up with to surpass this.
      • by Tmack (593755)

        I doubt Sun would come out with a system named Blue Steel. IBM, on the other hand, may be looking for something to call whatever they come up with to surpass this.

        IBM has already [slashdot.org] beaten this....(3 > 2)

        Tm

    • by sharkey (16670)
      Better than 'Higgins' and 'T.C.', that's for sure.
  • by PCM2 (4486) on Wednesday June 27, 2007 @04:18PM (#19668551) Homepage
    So Bechtolsheim says Sun has been "somewhat absent" from the supercomputer market in the last few years. OK, I'll bite. Exactly what markets has Sun been going gangbusters in since about 1999?

    Still, kudos to Sun, for real. Investors may get mad that Sun is full of terrific technology and solid R&D but can't seem to build the business model that will let Sun capitalize on it all. But from my perspective... God, that sounds almost refreshing, doesn't it?
    • So Bechtolsheim says Sun has been "somewhat absent" from the supercomputer market in the last few years. OK, I'll bite. Exactly what markets has Sun been going gangbusters in since about 1999?

      Funny, I was just wondering when exactly Sun was in the Super Computer market. Sun was a low-end Unix machine company that transformed itself into a high-end Unix server/mainframe company. (I remember drooling over the brand new 64-way specs of the E10000 Starfire.) AFAIK, they never even touched the domain of supercom

      • by PCM2 (4486) on Wednesday June 27, 2007 @04:29PM (#19668699) Homepage
        I think when Sun talks about supercomputing it's really talking about HPC/grid-type systems.

        FWIW, that's where Sun sees its future. Which makes sense. There's no point trying to compete with Linux for low-end applications (and by "low-end" I mean everything from desktops to simple Web-app servers). Sun has always been good at crafting products for that top 2% of customers who really, really need that high-availability or high-performance component that isn't going to make a difference for the other 98%. And Sun can charge for them.
        • Sun has always been good at crafting products for that top 2% of customers who really, really need that high-availability or high-performance component that isn't going to make a difference for the other 98%. And Sun can charge for them.

          Except that, in the long run, the platform of choice for the HPC/high-availability/high-high-peformance market is turning out to be Linux, thanks to IBM and it's HPC business. Ever wonder exactly why IBM contributed NUMA, SMP, numerous networking improvements and JFS? You

          • Re: (Score:3, Interesting)

            by PCM2 (4486)

            Except that, in the long run, the platform of choice for the HPC/high-availability/high-high-peformance market is turning out to be Linux, thanks to IBM and it's HPC business.

            Ha, well, yes, there is that. Only I wouldn't lie it completely at IBM's feet. After all, IBM is pretty much the only mainframe vendor still around. They have a vested interest in selling that kind of supercomputer, even though they've obviously seen the writing on the wall for their mainframe business.

            Outside the commercial secto

            • by Wolfrider (856)
              --Don't forget Amdahl... Altho they're a pretty distant 2nd, in my mind.

              --Linux-on-mainframe is just waiting in the wings rt now; they're not really "pushing" it yet.
        • Re: (Score:3, Interesting)

          by billcopc (196330)
          I second that. Sun's product line is very aggressive for SMB gear. They don't stray much into the supercomputing arena because it's a whole other ballgame, but for high-end "common" servers and workstations, they offer some pretty serious bang for the buck. In that light, they compete in the same segments as Dell's Poweredge line, or HP's Proliant. Medium iron as opposed to big iron. Server gear for the rest of us who aren't on the Fortune 50 :)

          I fell in love with Sun when I first laid my hands on a Su
        • Sun has always been good at crafting products for that top 2% of customers who really, really need that high-availability or high-performance component that isn't going to make a difference for the other 98%. And Sun can charge for them.

          Let's rephrase a tad. Sun has been good at SELLING products for that top 2% of customers who really, really need that high-availability or high-performance component that isn't going to make a difference for the other 98%. And Sun REALLY REALLY charges for them.

          Sun is in t

    • Re: (Score:3, Informative)

      by iggymanz (596061)
      Exactly what markets has Sun been going gangbusters in since about 1999?

      why the Java market, it's everywhere. of course, Sun hasn't made a thin dime off of it, but the market sure went for it.
      • by Nazlfrag (1035012)
        If you can't see the benefit for Sun to have widespread cross platform software available, you aren't looking.
        • by Wolfrider (856)
          --Yep. Mindshare is key. (Altho i wish vmware would get behind this is well, and release a completely free linux live-cd with Player.)
        • by iggymanz (596061)
          if you can't see the value of Sun turning their lack of profitability around before they go under you aren't looking.
          • by kaffiene (38781)
            Not to perturb you with actual facts, but Sun *are* actually making a profit.

            This, after the strategy of open sourcing all the family jewels has only been running for the last year. That's quite a turn around in quite a short period of time.
    • by TheRaven64 (641858) on Wednesday June 27, 2007 @05:00PM (#19668979) Journal

      Exactly what markets has Sun been going gangbusters in since about 1999?
      Web app servers. For a lot of web app type workloads, the T1 blows everything away in terms of power per watt and power per dollar.
      • by afidel (530433) on Wednesday June 27, 2007 @09:23PM (#19671197)
        Yep. We are implementing a datawarehouse app and after seeing the poor performance we are getting out of Oracle on Windows for our ERP product we are researching different solutions. Right now Sun is in the lead. They are about 20% less than our Windows solution for the overall solution and we expect it to be more scalable as we will only be partially populating the DB server out of the gate. They claim they can do with 4 middle tier boxes what our Windows solution provider has speced 13 Windows boxes for. I can't wait to see the results of the bakeoff. I'm primarily a Windows/Citrix guy but I've admin'd Solaris and Linux in past and I won't ming keeping my Solaris skills up to date if it's the best solution =)
    • by Anonymous McCartneyf (1037584) on Wednesday June 27, 2007 @05:09PM (#19669081) Homepage Journal
      Sun is responsible for OpenOffice. I expect OpenOffice to become the premiere office suite on PCs of all sorts any day now.
  • IBM Blue Gene/P (Score:4, Informative)

    by xzvf (924443) on Wednesday June 27, 2007 @04:19PM (#19668553)
    IBM Blue Gene/P update slated to run at 3 petaflops.
    • Re: (Score:3, Interesting)

      by flaming-opus (8186)
      yeah, there are a lot of systems out there with theoretical limits of several petaflops. Cray, IBM, NEC, even SGI have systems that could theoretically hit several PFlops if you had enough money.

      I'm waiting to see a customer actually purchase one, and for it to be installed, and actually running customer code, before I really care.
      • I'm waiting to see a customer actually purchase one, and for it to be installed, and actually running customer code, before I really care.
        It's unlikely to happen. Most of the people in a market for this kind of hardware have some serious legal restrictions preventing anyone mentioning them as a customer. Ever wonder how SGI still exists?
        • That's not actually true.
          Everyone in the industry knows who the customers are. Some of the specifics are secret, and what codes are actually run on the systems is not known. Really huge machines do have to be manufactured, and pushed through Q/A before being sent to a customer, even a classified customer. The on-site installation guys have security clearance, but no HPC company has all of the manufacturing and Q/A people cleared. If the secret-sites were buying fastest-in-the-world sized computers, people w
      • by xordos (631109)
        Here is the list of computers in use: http://www.top500.org/lists/2007/06/ [top500.org] and first SPARC is #178 http://www.top500.org/system/7522/ [top500.org]
    • by Life2Short (593815) on Wednesday June 27, 2007 @05:00PM (#19668989)
      Meh... In 15 years the thing will wind up as baby furniture [gizmodo.com] with kid puke on it anyway.
      • by AbRASiON (589899) *
        The replies to that post are fantastic.
        The couple who decked out that kids room, while thoughtful and probably quite intelligent are also dorks (not the good kind!) that poor kid is totally going to rebel just as the comments say.
    • Re: (Score:1, Insightful)

      by Anonymous Coward
      no, IBM hasn't built a production system yet, they haven't even achieved 3PF or even 1. They did the same thing they always do, benchmark a rack and extrapolate (in the most ideal and ridiculous situation) to an installation like 126 racks (I don't care who you are, that's a lot of freakin rack). Sun is doing the same thing, they don't have a 3k port switch, they're just saying they'll be releasing one. Look for an installation using one or an SKU to order.
    • Any likely costs 3 times more with 10x more cost in IBM global services..
    • by Ilgaz (86384) *

      IBM Blue Gene/P update slated to run at 3 petaflops.
      I heard on some IBM mainframes, you buy extra CPU and while you wait for engineer coming with IBM CPU boxes, guy comes and enters couple of passwords, mainframe enables the CPU it already has but disabled!

      Now imagine that on Blue Gene :)
      • by mikael (484)
        That's quite believable - with some customer locations and applications, it costs more to ship out a CPU module, fly out a field engineer, put the system on standby, insert the CPU module, test the system, reinstate the system into active use, and certify everything is working than it does to have the CPU module installed in the system but logically disabled, and activated by password.
  • And yet... (Score:5, Funny)

    by Anonymous Coward on Wednesday June 27, 2007 @04:19PM (#19668555)
    Java apps still take 3 minutes to start up on it.
  • by Bobfrankly1 (1043848) on Wednesday June 27, 2007 @04:19PM (#19668567)
    Well if they just move the petaflops out of the way before it gets there, they won't have to worry about it hitting them.
    -
    In Soviet Russia, TFA reads you!
  • by CompMD (522020) on Wednesday June 27, 2007 @04:21PM (#19668589)
    Please tell me the first production system will be named "Enterprise." There have to be enough people that will work on it that will be proponents of this.
  • Great units (Score:5, Funny)

    by Life700MB (930032) on Wednesday June 27, 2007 @04:21PM (#19668591)

    We are looking at a factor-of-three improvement over the current best system at an equal number of nodes

    Whoa, slowdown boy, just tell us how many laptop-miles of power this machine has!


    --
    Great hosting [dreamhost.com] 200GB Storage, 2_TB_ bandwidth, php, mysql, ssh, $7.95
  • by drspliff (652992) <harry.roberts@NOSPAM.midnight-labs.org> on Wednesday June 27, 2007 @04:24PM (#19668635)
    With Sun's slow improvements in the multi-core arena (the T1 & T2 systems) and their low power requirements their probably in for a good run at the top 10-20 spots on the Top 500 list.

    Consider 500 top-end T2 systems hooked up to some very fast switching hardware and you're performance per wattage ratio are going to be very persuasive to those running big data centers, although with the T1 systems the only thing which stopped us adopting them was the shared FPU (telephony codec transcoding sucks on them).

    Could we see suns equivalent of IBM's BlueGene system appearing next year? I definitely think so :)
    • Actually Suns offering is a 4-socket opteron blade system.

      Though the T2 does offer dramatically improved floating-point performance, as compared to the T1, I've seen no evidence that it out-performs better than quad-core opterons, on the sort of HPC workloads needed for these sorts of systems. The T2 is designed for transaction processing, which has very different needs.

      The main reason I don't think Sun would sell T2 for supercompute, is that they haven't mentioned it at all, and t's the sort of thing they
    • by Kristoph (242780) on Wednesday June 27, 2007 @04:40PM (#19668805)
      If you RTFA you will note that, actually, this particular system is built around the Barcelona architecture (from AMD). It remains to be seen if T2 and later on Rock will really be competative against AMD and Intel.

      ]{
  • 3456 = 2^7 * 3^3, so it's not a completely arbitrary number.
  • by athloi (1075845) on Wednesday June 27, 2007 @04:27PM (#19668677) Homepage Journal
    Back in the 1980s, what distinguished the Amiga (and later Steve Job's excellent NeXT) was the ability to split data among multiple co-processors and pipe it quickly around the motherboard, eliminating bottleneck and liberating the processor. Now in the PC world we're finally seeing this architecture recognized as new Intel chips tout their front-side bus and cache more than sheer increase in speed.

    This SUN machine is a bigger-scale example of the same. It uses AMD Barcelona chips, and derives its power from internally routing data more efficiently than (most of) its competitors. It seems that in the Moore's-law endgame, what makes the chip a star performer is the surrounding components and their engineering for efficiency.

    This will be better for geeks, as it makes the skill of efficient design come back into play after years of "bigger is better." Now if it just extends to software as well, we'll all benefit...
    • by kpharmer (452893) on Wednesday June 27, 2007 @05:07PM (#19669059)
      > in the PC world we're finally seeing this architecture recognized as new Intel chips tout their front-side bus and cache
      > more than sheer increase in speed.
      > This SUN machine is a bigger-scale example of the same.

      No, not really - parallelism has of course been around forever. But its application for high performance computing has been constantly demonstrated on large servers for the past 15 years. This is back when parallelism on intel hardware might at most have meant two cpus. And that was rare.

      MPP (Massively Parallel Processors) systems like Teradata and IBM's SP2 (aka DeepBlue - that defeated Kasparov at chess) successfully demonstrated great performance for the dollar back around 1994-1995 or so. These were originally designed around data mining and math computations - but found most of their sales in data warehousing. Meanwhile, CRAY was complaining that not all problems were good candidates for this kind of more cost-efficient hardware.

      By 1998 you could put db2 or informix on a hundred-node SP2, each node consisting of an eight-way SMP, each with its own dedicated storage. Queries on that old system were lightning fast compared to most other options. I worked directly on SP2s and worked with a team that has a 128-node one. Oracle & Sun eventually ecliped these solutions with massive SMPs. But much of that was more due to Informix's financial issues than technical merit - since Informix and DB2 (and of course Teradata) on MPPSs easily out-scaled oracle on SMPs. The SMPs were easier to adapt to application design changes, but the MPPs were easier to grow indefinitely large.

      These newer solutions are just more of the same thing - you've still got the same challenges in:
          - tons of OS and application images that must be consistent
          - node communication bandwidth (major selling feature of all these solutions are proprietary internal networks)
          - failover (how do nodes failover, especially if they have any dedicated resources)
          - scheduling (how do jobs get assigned to nodes)

      So, they're much bigger and faster than 10-15 years ago - and I'm sure there's got to be some cool innovation going on under the hood. But nothing looks fundamentally different from then. And nothing here has been inherited from the pc world.
    • Re: (Score:3, Informative)

      by fm6 (162816)

      It uses AMD Barcelona chips...
      AMD chips are an option. Blades come in AMD [sun.com], SPARC [sun.com], and Intel [sun.com] flavors.

    • The same argument was made for the Commodore 64.

      Do you really know what you're talking about? Can you be more specific about the specific data paths between components and how the operating system controlled them or interfaced with them?

      I mean, for example, if you wanted to make a video playback or slideshow program could you get the hard disk controller to blit directly to the screen bypassing main memory, or perhaps over the system bus, or a secondary channel? (I doubt it). Or could you describe a similar
      • by Nazlfrag (1035012)
        Well basically you had a set of coprocessors Agnus, Paula and Denise.

        Agnus was the memory controller chip (like the Sun chip), and had the blitter and copper which had independent access to the I/O bus and main+video memory (well video memory for the screen you were using could be anywhere in ram, allowing simple double buffering etc.) The copper has a simplified instruction set (move, wait and skip) and was tied to raster timings and good for screen effects while the blitter did blitting, and you could per

    • by chthon (580889)

      It has always been Intel's POV that they should be the sole deliverer of silicon for PC's.

      The reason for introducing MMX was the following. In 1995/1996 Philips Semiconductor was working on a project for creating a multi-media chip, which would have been an addition to the PC architecture. Intel did not like this and introduced MMX, so that they could say that such additional hardware was not necessary.

  • 3,456 (Score:5, Funny)

    by Nom du Keyboard (633989) on Wednesday June 27, 2007 @04:28PM (#19668693)
    3,456 ports. Now there's a non-computer number if I've ever seen one. It looks like someone asked, "And just how many ports do we need to be competitive," and someone else just started hitting the number keys in sequence across the top of the keyboard, starting at "3", until either Marketing was happy, or the engineer in charge fainted.
    • Re: (Score:3, Funny)

      by jd (1658)
      Or they needed 640 ports for internal connectivity (or 640 ports, channel-bonded, for upstream connectivity). Personally, I think it's the manager's password, though.
    • Re:3,456 (Score:5, Informative)

      by flaming-opus (8186) on Wednesday June 27, 2007 @04:41PM (#19668817)
      Actually 3456 is 12 X 12 X 12 X 2. It's not actually a 3456 port router, it's a fat tree of 24-port router modules. Each rank 1 & 2 module has 12 ports down and 12 ports up. The rank 3 modules have 12 ports down, and 12 sidelink ports to one another. Thus you end up with a 3456 port, rank 3.5 fat tree all in one box.
      • Re:3,456 (Score:5, Informative)

        by flaming-opus (8186) on Wednesday June 27, 2007 @04:49PM (#19668887)
        I'll add:
        This is not an unusual arrangement for existing infinaband networks. The distinction is that they have all of these 864 switch modules in a cabinet, and the wiring is probably traces on a backplane, rather than flexible cables. This improves the reliability, reduces the cost, and makes it a whole lot easier to install. That may sound silly, but you're talking about 10,000 cables, each with endpoint connectors on each end. Even buying in bulk, that's a lot of money in cables.
      • All sacred geometry or even cockblockery aside, do you remember the 5, 6, 7 & 8's?

        Now, someone help me with trees. I, for one, am unclear how a tree is applicable to ports. If I have an Ethernet router with 5678 ports--perhaps this is even better than 3456 ports-- I don't want some ports to talk immediately to the root node of a 12-ary tree while other ports have to send data meandering through twigs, branches, and trunks. If a 12-ary tree is good, a 3456-ary tree is better. Better yet may be every one
    • When building multi-stage fat trees from 24-port switch chips you get weird numbers like 288 and 3456, but rest assured they aren't random.
  • by TheWoozle (984500) on Wednesday June 27, 2007 @04:34PM (#19668747)

    The linchpin in the system is the switch, the piece of hardware that conducts traffic between the servers, memory and data storage. Code-named Magnum, the switch...
    I know what you're thinking. Did I forward 65,535 packets or 65,536 packets? Well, to tell you the truth, in all the excitement I kinda lost track myself. But being as this is a Sun Microsystems Magnum, the most powerful switch in the world, and would blow your IP clean off, you've got to ask yourself a question: Do I feel lucky? Well, do ya, punk? Go ahead. Make my day.
  • the switch comes with 3,456 ports

    Sounds to me like a giant USB-Hub to create a Beowulf-cluster of these [slashdot.org].

    And before somebody asks: Yes, it DOES run Linux.
  • But can it do anything useful?
  • ...just to handle all bittorrent streams.
  • by flaming-opus (8186) on Wednesday June 27, 2007 @05:02PM (#19669005)
    Bechtolsheim compares 131,000 cores of Blue Gene/L to 131,000 cores of constellation, with the sun system offering 3 times the performance.

    This is hardly a fair comparison. IBM installed a 131,000 core BG/L 2 years ago, and it's been running customer code for more than a year. The sun system won't be built until late this year, and probably won't be running real customer code until this time next year. Furthermore, the BG/L machine is designed with a low-power node, assuming that a larger number of cores would be used. In IBM's older BG/L design, there are 2048 cores in a rack. Sun is packing 768 opteron cores in a rack. So a per square-meter measure gives IBM's 3 year old design only a 20% disadvantage to Sun's not-yet-released machine.

    All of that is moot, of course, as theoretical peak performance is a crappy way to measure supercomputer performance anyway. The opteron is a great processor, and infinaband is a decent, though not remarkable interconnect. I'd be a little concerned, were I to buy the sun solution, that the infinaband bandwidth is being shared by 16 processor cores. That's quite a bit less interconnect performance per processor than IBM's Blue Gene, power5, Cray's XT, or SGI's altix. There's certainly plenty of memory on each of these constellation blades. That said, there are a list of applications that perform very well on Blue Gene, and Sun has a lot of ground to make up in terms of OS, software, and establishing a relationship with the HPC customers.

    It's nice to have more options, however.
    • by Rhys (96510)
      A lot of ground to make up... like porting Linux to the AMD processors (or Intel, Sparc as mentioned elsewhere)? Look at the top 500 list's operating system breakdown and you'll find a remarkable amount of Linux.

      I'd like to agree to be concerned about the bandwidth, but I'm not sure I can be. Looking at my friendly local supercomputer over the last year, I see the most processor-hours used by 64 proc jobs (33%), followed by 32 proc jobs (27%). If jobs keep fairly consistent on the number of processor cores
  • Code-named Magnum, the switch comes with 3,456 ports
    Sun: just one look?! JUST ONE LOOK?!
  • It's not the number of petaflops, it's how you use it.

    Geez, Sun and IBM getting into a prick-waving contest.
  • Overkill? (Score:2, Funny)

    by Red Mage 13 (791885)
    640k should be enough.
  • by sonoronos (610381) on Wednesday June 27, 2007 @06:15PM (#19669665)
    Have you ever wondered if there was more to life, other than being really, really, ridiculously high-performance?
    • Have you ever wondered if there was more to life, other than being really, really, ridiculously high-performance?
      Of course there's more.

      -Bad car analogies
      -Outdated "profit!" jokes
      -Dupes

      etc.
    • by pimpimpim (811140)
      For those doing research, it will mean that they can either increase the system size by ten, just to annoy other research groups that don't have this access, or that they will have to work much more hours because the calculations will be finished so much faster that they don't have time to write on the articles while waiting for the computer to finish.

      In practice, it means that you can be less careful when preparing your startfile. Imagine screwing up one input parameter or line of code 30 years ago, that

  • Are they entering with a new grid? Or just entering the market as a switch provider. Either way it sounds like they've done some good work. But if it's for an actual grid and not just switches what CPU's are they using? I use to love Sparc back in the day along with Alpha, but they both seem to have more or less died in the last 4-5 years. Did Sparc ever get past 2ghz?

    The switches do sound awesome though, will give infiniband a run for their money.

    • by MidKnight (19766)
      You may want to read the friggin' article. [zdnetasia.com] After you do, you might know that ...
      • The switch is the center of the system. When you buy the switch, you also buy cabinets full of blade systems. Sun won't sell you the switch as a stand-alone component.
      • The blades in cabinets can be SPARC, AMD, or Intel; Sun will support them all. The first one will be delivered to the Texas Advanced Computing Center, and will use Barcelona chips from AMD because it has better FP performance
      • The switch is built around Infini
  • This will be when and only when the headline actually reads, "Sun Super Computer Hits 2 Petaflops".
  • The police will finally have something they can process their stored petafiles with! Two at a time no less!
  • by damacus (827187)
    Wake me up when they've released the galaxy class system. Binary is so passe. All the cool kids use trinary systems. Viva la trek!
  • So finaly I'll be able to solve all those pesky NP-Hard problems.

  • "...and two petaflops should be enough for anybody!"

Be careful when a loop exits to the same place from side and bottom.

Working...