Slashdot Mirror


Cray XD1 Now Available

cyngus writes "Cray announced the availability of their XD1 systems. Each XD1 chassis has up to 12 AMD Operton processors. Up to 12 chassis can be clustered together in a rack. The XD1 uses Cray RapidArray Interconnect technology, based on HyperTransport, for high bandwidth and low latency communications between processors and chassises. The XD1 also has a handful of other technologies aimed at the HPC market, including Xilinx FPGAs, communications accelerators, etc."

200 comments

  1. Not the top end by Oculus+Habent · · Score: 4, Funny

    It would take sixty racks of these to best the Earth Simulator's theoretical peak; more than 60% more processors.

    Still, if they need someone to, uh, test one...

    --
    That what was all this school was for... to teach us how to solve our own problems. -- janeowit
    1. Re:Not the top end by LoudMusic · · Score: 2, Interesting

      It would take sixty racks of these to best the Earth Simulator's theoretical peak; more than 60% more processors.

      Still, if they need someone to, uh, test one...


      Interesting numbers. Also to note, NEC's Earth Simulator is now nearly three years old - the Cray XD1 is made with modern AMDs.

      I guess there's no getting around it. For the time being our really fast computers will just be fucking huge. Oh, and call NEC if you want a big computer. (:

      --
      No sig for you. YOU GET NO SIG!
    2. Re:Not the top end by Anonymous Coward · · Score: 1, Interesting

      I guess there's no getting around it. For the time being our really fast computers will just be fucking huge. Oh, and call NEC if you want a big computer. (:

      Actually:

      Blue Gene

      Let's see what happens come Supercomputing '04... good things can come in small sizes. ;)

    3. Re:Not the top end by Rasta+Prefect · · Score: 2, Interesting
      It would take sixty racks of these to best the Earth Simulator's theoretical peak; more than 60% more processors.

      Earth Simulator uses vector processors. If you want a comparable Cray system, you should be looking at the X1 which is also a vector processor. Incidentally, the X1's silicon runs so hot they use evaporative florinert cooling instead of a straight liquid - the florinert is heated to just under the evaporation point and sprayed onto the processor so that the phase change will remove more heat than just immersion.

      --
      Why?
    4. Re:Not the top end by visgoth · · Score: 4, Insightful
      I'd say that the fastest computers will always be fucking huge. If engineers could somehow magically fit the total power of the Earth Simulator into a single 1u chassis, people would still cluster a few hundred of them together. There's no such thing as enough processing power, as people always find a way to utilize it.

      (commence snide comments about the next windows release... now :) )

      --
      My patience is infinite, my time is not.
    5. Re:Not the top end by flaming-opus · · Score: 2, Interesting

      ISR (Isothermal Systems Research, Inc.) and cray have cross licensed patents on this technology. I don't know if ISR plans on productizing this or not.

      I imagine that this is extremely expensive stuff to do. Since a cray can charge $40,000 per processor for the X1, they can get away with a $700 cooler. Not so easy on a PC.

    6. Re:Not the top end by bobbozzo · · Score: 2, Interesting

      You shouldn't need phase-change to cool a PC.

      Just circulate the stuff and run it through a radiator.

      BTW, HP was recently researching cooling chips with inkjet nozzles spraying a coolant which evaporates easily onto the CPU.

      --
      Nothing to see here; Move along.
    7. Re:Not the top end by gstoddart · · Score: 1
      I'd say that the fastest computers will always be fucking huge ...

      snip

      commence snide comments about the next windows release... now :)


      I for one welcome our new fucking-huge-windows overlords. :-P

      --
      Lost at C:>. Found at C.
    8. Re:Not the top end by LoudMusic · · Score: 1

      Actually:

      Blue Gene

      Let's see what happens come Supercomputing '04... good things can come in small sizes. ;)


      From the linked site ...

      with a sustained speed of 11.68 teraflops and a peak speed of 16 teraflops, uses more than 8,000 PowerPC processors packed into just four refrigerator-sized racks.

      So lets say their "refridgerator sized racks" are about the size of 1.5 standard racks, and you'd need about 3x the amount of flops they're getting currently, so that's 36 racks? Not too shabby. Almost half the space of the Cray AMD setup. I guess I didn't realize it was so nicely packaged (:

      --
      No sig for you. YOU GET NO SIG!
    9. Re:Not the top end by LoudMusic · · Score: 1

      I'd say that the fastest computers will always be fucking huge. If engineers could somehow magically fit the total power of the Earth Simulator into a single 1u chassis, people would still cluster a few hundred of them together. There's no such thing as enough processing power, as people always find a way to utilize it.

      Hmmm ... good point. Sometimes I slip into "Apple Mind" and forget that technology has progressed and think that my workstation is a super computer (referance to the G4 'your own personal supercomputer' ad campaign). No matter how much processing power we can cram into a small space, someone will still stick a bunch of them together and run dnetc on it (:

      --
      No sig for you. YOU GET NO SIG!
    10. Re:Not the top end by Glonoinha · · Score: 1

      So what the hell do we do with it? Granted it is a little bigger, faster than the 10GHZ to 30GHz cluster I'm asking about, but after you work out the details it is just a matter of scale.

      This seems as good a place as any to ask that question (see my sig for details.)

      --
      Glonoinha the MebiByte Slayer
    11. Re:Not the top end by foobsr · · Score: 1

      I'd say that the fastest computers will always be fucking huge.

      Make earth the computer - or is it in the end ???

      Huge and clustering will then at least take a little while longer :) Hmm, what about wormhole (WoFi) connected planets? This will again lead to an appled Planetport. I digress.

      CC.

      --
      TaijiQuan (Huang, 5 loosenings)
  2. long time no news... by mirko · · Score: 4, Insightful

    Since they had been bought by SGI, I've actually been wondering whether they would make me dream again.

    --
    Trolling using another account since 2005.
    1. Re:long time no news... by cyngus · · Score: 5, Interesting

      SGI does not own CRAY. They did buy them back in 1996. SGI sold its Cray unit in 2000 to Tera Computer.

    2. Re:long time no news... by Anonymous Coward · · Score: 1, Interesting

      well I think since the advent of home computer beowulf clusters it has take the awe out of huge supercomputers. Before it wasn't technically possible to scale a system to such high performance, but now with a bit of cash you can go buy OTS G%s and make a top500 computer. It is taking a bit more to inspire dreams with simply performance specs.

    3. Re:long time no news... by robslimo · · Score: 2, Insightful

      I think this is evidence that they are headed in the right direction. Rather than completely custom hardware all the way down to the processors, they've positioned themselves more competitively by using the CPU design power of a major business and consumer supplier as well as applying their own special design talents to other hardware areas that are not served by the business/consumer vendors.

      I too was worried that Cray would completely disappear if they continued to pursue the expensive and anachronistic supercomputer design model.

    4. Re:long time no news... by chegosaurus · · Score: 1

      Didn't Sun buy the best part of Cray, hence the E10k, or am I utterly wrong? (Again.)

    5. Re:long time no news... by soyle · · Score: 2, Informative
      Didn't Sun buy the best part of Cray, hence the E10k, or am I utterly wrong?


      I forget where I snipped this from, but here goes:


      The E10000 is a Celerity product. Celerity was an independent Unix box maker back in the 80's with their own processor architecture. Celerity went bust trying to bring a "minisupercomputer" version of the architecture to market in about 1987 (33 MHz, whoo hoo!). The assets and technology of Celerity along with the design team in San Diego were acquired by Floating Point Systems (FPS). FPS brought the system to market and made the transition to a SPARC based architecture (66 MHz) before going bust. The assets and technology of FPS along with the design team in San Diego and now the manufacturing team in Beaverton were acquired by Cray. Cray did a couple of turns of the crank on the FPS product and sold it as a "business supercomputer". When Cray was acquired by SGI, SGI wanted no part of the SPARC business and sold (yes, again) the San Diego design team (and I think the Beaverton group) to Sun who finally brought a SUCCESSFUL product to market with the E10000.

      But it's still the same core team down in San Diego, so I like to think of the E10000 as being a Celerity product.
    6. Re:long time no news... by Anonymous Coward · · Score: 0

      I bet your just waiting for the Trivial Pursuit question about this.

    7. Re:long time no news... by Bingo+Foo · · Score: 1

      No, he's waiting for a question on distinguishing possessive pronouns from contractions.

      --
      taken! (by Davidleeroth) Thanks Bingo Foo!
    8. Re:long time no news... by Anonymous Coward · · Score: 0

      Thank you for pointing out my mistake.

      Please replace "your" with "you are" or "you're".

      I usually try to be pretty careful with post, but with that one being so short, I rushed it through.

  3. Find the Fake Ad? by JackBuckley · · Score: 0

    These press releases ripped right off of PR newswire and posted to Slashdot always remind me of the "find the fake ad" section of Games magazine. "Find the fake news story planted on Slashdot to generate `buzz'". Bah. I hate marketing. J

    1. Re:Find the Fake Ad? by cyngus · · Score: 2, Funny

      Hey jack nuts, I posted this. I like Cray, because I think companies that put a lot of thought into their product and make great ones deserve a cheering section. Of course you're a BC kid, so I'll forgive you, we (BU) spanks you enough in hockey to let you have a shot here and there.

    2. Re:Find the Fake Ad? by JackBuckley · · Score: 1
      Hey, nothing personal--I have a soft spot in my heart for Crays, too. And this story is a *lot* better than yesterday's topper: Gamespot has pictures of the box that Nintendo is going to ship the DS in!!

      Obviously I pissed somebody off, cause I got modded down as "overrated." I just get worn down by the endless parade of PR trumpeted as news in the mainstream media. And sorry about your misfortune in attending BU.

  4. Cray by dnno · · Score: 0

    I remember reading about Dr. Cray (doctor i think). Is he dead I keep on thinking so anyone know? (BTW: That is one sweet arse 'puter) ::drools::

    --
    feh, lots of things are pointless, this one too
    1. Re:Cray by goneutt · · Score: 3, Informative

      Dr Cray was killed in 1995(in think thats the year) when his SUV was T-boned by a short car and rolled over.

      I only know this from a Sci-Am article on using supercomputers to predict crash situation.

      --
      Bacardi + slashdot = negative karma.
    2. Re:Cray by Handbrewer · · Score: 3, Informative

      Hes dead.
      http://en.wikipedia.org/wiki/Seymour_Cray
      Seymour Roger Cray (September 28, 1925 - October 5, 1996) was a supercomputer architect who founded the company Cray Research. For about 30 years, the short answer to the question "What company makes the fastest computer?" was "Wherever Seymour Cray is working now."

    3. Re:Cray by dangogh · · Score: 1

      yes -- Seymour Cray died in an auto accident in Colorado Springs in 1996.. I googled "Seymour Cray" for this: http://ei.cs.vt.edu/~history/Cray.Pepper.html

    4. Re:Cray by Twirlip+of+the+Mists · · Score: 1

      My all-time favorite Semour Cray quote is this one, possibly apocryphal: "A supercomputer is a tool for turning compute-bound problems into I/O-bound problems."

      --

      I write in my journal
  5. Oh, yea? by Anonymous Coward · · Score: 0

    Mr "would you rather have an ox plow a field, or 1024 chickens"' company making clusters now, eh? Yes, I know the man Cray is dead.

  6. But does it run Linux? by Doc+Ruby · · Score: 1

    Maybe! Linux runs on the MicroBlaze softCPU on Xilinx FPGAs, as a derivitave of uCLinux. Care to port it to this new hot hardware?

    --

    --
    make install -not war

    1. Re:But does it run Linux? by nenolod · · Score: 2, Informative

      Being based on Opteron, any x86 software will run on it. Maybe without all the bells and whistles though. But Openmosix can solve most of those problems.

    2. Re:But does it run Linux? by bsd4me · · Score: 2, Informative

      Having Linux or any other OS (or even CPU type functions) on the FPGA would be a waste of gates. The gates would be better spent for specialized vector operations, such as an FFT or crypto engine.

      --

      (S(SKK)(SKK))(S(SKK)(SKK))

    3. Re:But does it run Linux? by Doc+Ruby · · Score: 2, Insightful

      Not when you want to run existing Linux apps on those gates, without rewriting them (which would be a waste of programmer time, much more valuable). And running Linux apps on those gates is a first step in porting them to native FPGA netlists. This is the way to leverage existing apps to increase the utility of FPGAs, by running useful apps on them, and optimizing to native parallel execution. That way we don't waste either gates OR programmer time, not to mention every other resource in the chain.

      --

      --
      make install -not war

    4. Re:But does it run Linux? by fpga_guy · · Score: 1
      Maybe! Linux runs on the MicroBlaze softCPU on Xilinx FPGAs, as a derivitave of uCLinux. Care to port it to this new hot hardware?

      Hey - that's me! :)

      In answer to the child of parent (sibling?) the purpose of putting a softcore processor and Linux on an FPGA is to leverage all the things that Linux does so well, and also give access to the things that FPGAs do well.

      Ever tried to implement a TCP/IP stack and web server in pure logic? It's been done, but it's pain all the way. With Linux, it's there already, for free.

      Our research focusses on efficient ways of integrating the custom hardware capabilities within the operating system context.

      And, in case you were wondering, we are imagining a Beowulf cluster of these...

  7. Re:Just imagine... by Anonymous Coward · · Score: 0

    That would almost be enough computing power to run spell check!

    Or to install Gentoo in under 12 hours...

  8. Re:Banning Fun by Doc+Ruby · · Score: 2, Funny

    " this is a chance to hunt them down ". Go get 'em, tiger!

    --

    --
    make install -not war

  9. Sheesh...what happened to Cray? by ferkelparade · · Score: 5, Funny

    A Cray is not a true Cray unless it can be used as a stylish sofa :p

    --
    frotz grue
    1. Re:Sheesh...what happened to Cray? by 14erCleaner · · Score: 3, Funny

      That must be why Cray Computer failed, although the Cray-3 did make an attractive armrest.

      --
      Have you read my blog lately?
    2. Re:Sheesh...what happened to Cray? by Anonymous Coward · · Score: 0

      Speaking of true Crays, isn't this hive-design completely against the design philosophy that characterized and defined the original Crays against other supercomputers?

    3. Re:Sheesh...what happened to Cray? by soyle · · Score: 1
      A Cray is not a true Cray unless it can be used as a stylish sofa

      Which reminds me: I have always wondered what it would have been like to sit on one of those? Were there any mechanical parts inside the main tower making obnoxious amounts of noise?
    4. Re:Sheesh...what happened to Cray? by Linker3000 · · Score: 2, Informative

      Dunno, but I bet you got a warm tootsie because the seat held part of the cooling system and power supply.

      --
      AT&ROFLMAO
    5. Re:Sheesh...what happened to Cray? by ferkelparade · · Score: 2, Informative

      The only Cray-I I've ever seen up close is the one at Deutsches Museum in Munich. It's actually quite comfortable to sit on, and the insides (at least right behind the panels, some of which have been replaced with clear plastic sheets) seem to contain nothing mecchanical (here's a photo of the museum's Cray). Might get a bit warm if you sit on tere while it's powered on, though :p

      --
      frotz grue
    6. Re:Sheesh...what happened to Cray? by Anonymous Coward · · Score: 1, Interesting

      The Cray-1 and X-MP machines had soft padded versions and the Y-MP had hard plastic versions. Most of the really loud stuff was in the motor generator room off to the side. The ambient noise in the typical larger computer room was louder than anything coming from the Cray. The soft padded versions weren't the most comfortable things to sleep on, but they did in a pinch very late at night.

    7. Re:Sheesh...what happened to Cray? by Seltsam · · Score: 1

      I've sat on the very first Cray 1 (in a small computer museum in Chippewa Falls, WI). They aren't very padded after all those years.

      The Cray employee who gave us the tour, Bruce Serling (I think), told us he used to sleep on the first Cray 1 after a long day of work.

    8. Re:Sheesh...what happened to Cray? by Seltsam · · Score: 1

      ...and the Cray 2 made a good bartender turret.

    9. Re:Sheesh...what happened to Cray? by StoatBringer · · Score: 0

      I used to work at British Aerospace many years ago, and once got to go and see their lovely red Cray 1.

      Unfortunately, I was stopped from sitting on it as the machine-room guys said that even though it looked rather comfy it wasn't really designed for that sort of thing.

      Can you get them on eBay yet? I really, really want to sit on one.

      --
      Cress, cress, lovely lovely cress
  10. Operton? by ari_j · · Score: 3, Funny

    So, is the Operton more or less powerful than the Opteron?

    Also, mandatory: imagine a Beowulf cluster of these.

    1. Re:Operton? by nomadic · · Score: 3, Funny

      So, is the Operton more or less powerful than the Opteron?

      Also, mandatory: imagine a Beowulf cluster of these.


      Don't you mean Bewoulf?

    2. Re:Operton? by Anonymous Coward · · Score: 0

      Modded "1, Offtopic" ??

      Is there no mod "+5, The editors aren't" ?

      It's amusing to laugh at them, but some stories actually become meaningless with all the errors.

    3. Re:Operton? by Kehvarl · · Score: 1

      Be-Woulf?

      does it have dasblinkenlights?

    4. Re:Operton? by niteice · · Score: 1

      /me gets inspiration

      Imagine a beowulf cluster of blinkenlights!

      --
      ROMANES EUNT DOMUS
    5. Re:Operton? by Kehvarl · · Score: 2, Funny

      It could outshine my Tolkein Ring.

  11. just cause someone has to by FienX · · Score: 0, Redundant

    just imagine a beowu....nevermind

  12. Re:Just imagine... by raider_red · · Score: 1, Flamebait

    You can laugh, but you'll probably need that much to run the next iteration of Windows.

    --
    It's good to use your head, but not as a battering ram.
  13. Re:Cray Yes, he's dead by bobalu · · Score: 0, Redundant

    He died in a traffic accident. Some fool took him out. I dunno why they'd let a national treasure actually drive himself to work at that age, but there ya go. He was 72 I believe.

    --
    The revolution will NOT be televised.
  14. Does the XD1 give the illusion of shared memory? by Anonymous Coward · · Score: 3, Interesting

    I've heard conflicting reports on this - reading Cray's own literature, you see them say:

    "Tightly coupled to the AMD Opterons and switching fabric, [the RapidArray Communications Processors] handle memory to memory copies, global memory management, and system wide process synchronization, freeing..."

    (Emphasis mine)

    Does this mean the HT links give the OS the view of a single-system for each chassis? (Or rack, even?) Ie, can I utilize a single processor out of those 12 in a chassis, and access 96GB of RAM with that one process WITHOUT using MPI or rDMA?

  15. Find the non ad? by Doc+Ruby · · Score: 1

    Most computer journalism is PR. One of the reasons I like computing is that many of the ads are interesting. I read some of the paper magazines in the field mainly for their ads. It's a competitive marketplace of ideas, even though some are bad and wrong. There's very little news about tech that isn't announcements about products, even free ones. What's the alternative? _Nerd People Magazine_ (shudder)?

    --

    --
    make install -not war

  16. Cray doesn't do Clusters? by Edward+Ka-Spel · · Score: 3, Interesting

    I thought Cray was trying to convince the world that Clusters were not as good as true supercomputers, but this looks like a glorified cluster. In looking under the hood it appears to be just a collection of 2-way SMP Opterons with a superfast proprietary network backbone.

    And it's running Linux, if that matters to you

    1. Re:Cray doesn't do Clusters? by ari_j · · Score: 3, Interesting

      This is nothing compared to the 1024-way massively parallel computers Cray has built in the past. That's why they don't like clusters - to them, a cluster is an inferior solution in the light of massively parallel systems and, on the other end, vector supercomputers. What can you do with a cluster that you can't with one of these?

      Given the financial status of Cray, embracing clusters is just a common sense move, not necessarily an ideological one.

    2. Re:Cray doesn't do Clusters? by garethwi · · Score: 1

      You're right. The comment regarding clusters was 'Would you use two strong shire horses to plough a field, or 1000 rats?'.

    3. Re:Cray doesn't do Clusters? by Anonymous Coward · · Score: 0

      This is nothing compared to the 1024-way massively parallel computers Cray has built in the past. That's why they don't like clusters - to them, a cluster is an inferior solution in the light of massively parallel systems and, on the other end, vector supercomputers. What can you do with a cluster that you can't with one of these?

      Given the financial status of Cray, embracing clusters is just a common sense move, not necessarily an ideological one.


      And just one single 4U chassis in this new setup of theirs eclipses the sum total of all their previous supercomputers.

    4. Re:Cray doesn't do Clusters? by ianturton · · Score: 1
      It's not that different from the massively parallel machines. They were just DEC Alphas tied together with a fast interconnect.

      Ian

    5. Re:Cray doesn't do Clusters? by Dracolytch · · Score: 1

      And I believe the answer customers respond with is: Save lots, and lots of money. R&D, Purchase, and Maintenance.

      In the world of buisness, ideology often needs to take a back seat to good common sense. Personally, I feel it's to their credit. This way they can focus less on their last innovation, and more on the next one.

      ~D

      --
      This sig has been enciphered with a one-time pad. It could say almost anything.
    6. Re:Cray doesn't do Clusters? by tallganglyguy · · Score: 2, Insightful

      This is more than your typical cluster with a fast interconnect bolted on. I would rather picture it as something completely different with a glorified cluster bolted on. The true beauty of the system lies not only in the fast interconnect, but also in the 6 FPGAs that are onboard each chassis. The power of a properly implemented algorithm in hardware is a force to be dealt with. This setup by and far is not your typical cluster.

    7. Re:Cray doesn't do Clusters? by bhima · · Score: 1
      I suppose it depends on how well you can control the rats and how much you pay them.

      Seriously though, I some problems are well solved with clusters and some with vector processors. So I suppose Cray is trying to find out just how many fall into each camp... either that or how much their name is worth.

      Just thinking about that, I guess I would buy a mini Cray deskside thing if I had to replace my Onyx 2 but not a cluster as my application doesn't do well with them. Shame really, I like the Blue Gene/L.

      --
      Nothing in the world is more dangerous than sincere ignorance and conscientious stupidity.
    8. Re:Cray doesn't do Clusters? by raehl · · Score: 1

      What can you do with a cluster that you can't with one of these?

      Spend a few million on something else.

    9. Re:Cray doesn't do Clusters? by pkhuong · · Score: 1

      A vector processor is SIMD; a cluster is MIMD. MIMD is a superset of SIMD, but, _theoretically_, less effective. The advantage of SIMD is its simplicity, but when MIMD is effectively simpler to use than SIMD, vector processors are obviously only used when one has to.

      --
      Try Corewar @ www.koth.org - rec.games.corewar
    10. Re:Cray doesn't do Clusters? by Hoser+McMoose · · Score: 1

      In a way Cray DIDN'T do this cluster. It was actually a company called OctigaBay that created this design, but they were subsequently bought out by Cray early this year.

      Cray's own design for an Opteron-based system is their Strider/Red Storm design, which is a somewhat more traditional Cray-style system.

  17. Must.. get.. ridiculously.. powerful.. device.. by phyruxus · · Score: 5, Funny
    Dogbert: So, what does it do?

    Dilbert: I can compute many values of pi. Some people discuss areas of circles, but I'm doing something about it!

    --
    "A witty saying proves nothing." ~Voltaire
    "d'Oh!" ~Homer
  18. Interesting specs and density by sczimme · · Score: 4, Informative


    From the linked page:

    Highly modular, the Cray XD1 base unit is a chassis. Up to 12 chassis can be installed in a rack. Multirack configurations integrate hundreds of processors into a single system.

    Farther down the same page:

    The Cray XD1 compute subsystem is composed of 12 AMD Opteron(TM) 64-bit processors that run Linux and are organized as six 2-way SMPs to deliver 58 GFLOPs* per chassis. Finely tuned memory and I/O performance removes bottlenecks and maximizes processor performance.


    Wow - do the math: 696 GFLOPs per chassis. That's rather impressive.

    However, part of me is a bit saddened by seeing the Cray name attached to X86s. Yes, I felt the same thing with SGI, DEC, and Sun. Yes, I need to get over it and move on. :-)

    /occasionally misses the Y-MP

    --
    I want to drag this out as long as possible. Bring me my protractor.
    1. Re:Interesting specs and density by red_dragon · · Score: 1

      However, part of me is a bit saddened by seeing the Cray name attached to X86s. Yes, I felt the same thing with SGI, DEC, and Sun. Yes, I need to get over it and move on. :-)

      No need to worry -- if you want a Cray vector processor-based supercomputer, you can still buy one. You can also source the parts for your own Earth Simulator from Cray, as well.

      --
      In Soviet Russia, Jesus asks: "What Would You Do?"
    2. Re:Interesting specs and density by 14erCleaner · · Score: 2, Interesting
      However, part of me is a bit saddened by seeing the Cray name attached to X86s.

      Actually, in the year between crash of Cray Computer (in March 1995) and his death in an auto accident, Seymour Cray started a new company, SRC Computers, which still exists, and makes a parallel Pentium-based computer (which also incorporates custom hardware processing elements). I believe that this product is the same thing he was working on from the start of that company in 1996.

      --
      Have you read my blog lately?
    3. Re:Interesting specs and density by Epi-man · · Score: 1

      From your quote of the article:
      The Cray XD1 compute subsystem is composed of 12 AMD Opteron(TM) 64-bit processors that run Linux and are organized as six 2-way SMPs to deliver 58 GFLOPs* per chassis.

      Your comment:
      Wow - do the math: 696 GFLOPs per chassis. That's rather impressive.

      To get the 696 GFLOPS you need to have 12 chassis (a fully loaded system), so it isn't 696 GFLOPS/chassis, but /system. Sorry, just wanted to point it out.

    4. Re:Interesting specs and density by CommieOverlord · · Score: 1

      This product is the design of a BC, Canada startup called Octiga Bay. The team behind it were former telco switch/router engineers/execs. From Nortel, I believe. Nothing to do with SRC

    5. Re:Interesting specs and density by 14erCleaner · · Score: 1

      I understand, I was just pointing out that Seymour's company was planning to use Pentiums when he had his accident.

      --
      Have you read my blog lately?
    6. Re:Interesting specs and density by Glonoinha · · Score: 1

      Damn, how's that for a sign from God?
      (Yes, I'm probably going to hell for that one.)

      --
      Glonoinha the MebiByte Slayer
  19. If it were a girl robot. by Icegryphon · · Score: 0, Redundant

    I would Hit it!.

    1. Re:If it were a girl robot. by Anonymous Coward · · Score: 0

      That's "I'd hit it!".

      gb2/b/...

    2. Re:If it were a girl robot. by drinkypoo · · Score: 1

      This meme has been around so long that my great-great grandfather used to include it in his messages sent by carrier pigeon.

      --
      "You're right," Fisheye says. "I should have set it on 'whip' or 'chop.'"
    3. Re:If it were a girl robot. by Anonymous Coward · · Score: 0

      Please, the XD1 wouldn't even give you her phone number.

  20. Re:Does the XD1 give the illusion of shared memory by IWannaBeAnAC · · Score: 2, Informative

    Yes, I believe it is 12-way SMP. The memory is connected to the CPU's in a crossbar switch.

  21. Not chassises by Anonymous Coward · · Score: 1, Funny

    The plural of chassis is chassis. This Gollum thing on /. is going too far.

  22. XD1 announced sales by ruiner5000 · · Score: 2, Informative

    Cray has announced a lot of different sales of the XD1 the past couple of weeks. We have all the details here.

    --
    ignorance is bliss. googlefiberatx.com
  23. the difference by bmajik · · Score: 5, Informative

    the nec SX architecture uses these ridiculously huge custom vector processors to get performance (similar to the Cray 1, 2, XMP, YMP, etc design)

    this Cray is more like building MPPs off of scalar units (opterons) and doing some real innovation around the MPP interconnect. It's sort of off the shelf, yet not at the same time.

    The big thing here that kicks ass is the 6 FPGAs per chassis. If you can write a highly tuned software algorithm, there's a chance you can write a highly tuned peice of hardware, deploy that to the FPGA, and you've got an application specific hardware accelerator. 6 per chassis, infact. That's pretty cool, and its in some ways a HUGE innovation over having a dedicated vector unit (as was the cray1 design).

    the really interesting thing here is that these are essentially opterons running linux, with custom interconnect goo. The interconnect bypasses the PCI bus - its closer to the PE's than that.. their claim is that it attaches to the AMD hypertransport bus (the Proc -> Proc -> Mem bus for SMP AMD machines)

    --
    My opinions are my own, and do not necessarily represent those of my employer.
    1. Re:the difference by bobbozzo · · Score: 1

      Last I heard, Cray and AMD had made a custom Infiniband-Hypertransport bridge for this system.

      It would be nice if it were available on low-cost motherboards so anyone could build a similar system.

      --
      Nothing to see here; Move along.
    2. Re:the difference by Anonymous Coward · · Score: 0

      For clarity's sake... each vector processor of the Earth Simulator resides on a single chip, about 2cm by 2cm. Ridiculously huge they aren't...

    3. Re:the difference by Anonymous Coward · · Score: 0

      I was part of a small team that built the very first ever reconfigurable computing platform using Xilinx FPGA's. At the time, the Xilinx company engineers thought we were kind of nuts, not so anymore!

      Our project was killed primarily due to incompetence of upper management. Myself and the other lead engineer on the project left because of undelivered promises, lies, payments, bonuses, etc.

      Check out Kent's VIVA platform at www.starbridgesystems.com

      If you want to break some crypto...

  24. Where's the source code by Donny+Smith · · Score: 2, Insightful

    > Cray HPC-enhanced Linux, Kernel version 2.4.21

    I wonder what that means - Red Hat EL 3.0 with enhancements, or their own thing..

    Interconnect - I wonder how their proprietary interconnect compares to IB..

    File system - ext3? No cluster file system?

    1. Re:Where's the source code by Anonymous Coward · · Score: 0

      More importantly, I hope they are using enhancements from the 2.6 series. The Opterons are much, much better on 2.6 than on 2.4.

    2. Re:Where's the source code by Anonymous Coward · · Score: 0

      More importantly, I hope they are using enhancements from the 2.6 series. The Opterons are much, much better on 2.6 than on 2.4.

      No, they're using a bog standard 2.4 kernel compiled for i386 with math emulation built in and no SMP support.

      IT'S CRAY - HAS IT OCCURRED TO YOU THAT THEY MIGHT KNOW WHAT THEY'RE DOING?/b?

    3. Re:Where's the source code by Blackfire · · Score: 1

      I believe it is based on Suse. Someone please correct me if I'm wrong.

    4. Re:Where's the source code by Troy+Baer · · Score: 2, Interesting
      Actually, as far as I can tell, their interconnect is IB at the hardware level, but without a PCI bus between the HCA and the memory controller. That drops their latency by a couple microseconds, and it means they don't need PCI Express to get full bandwidth out of the network. The claim is that they use their own software stack in place of the VAPI stuff, and that they don't use MVAPICH from Ohio State like most IB sites do. I haven't had a chance to look at our XD1 (we've had 3 chassis' worth for almost a month) to see if that's true or not.

      Their story as far as storage is kinda lame; I think they're praying Lustre won't suck.

      --
      "My life's work has been to prompt others... and be forgotten." --Cyrano de Bergerac
    5. Re:Where's the source code by Spoing · · Score: 1
      1. I wonder what that means - Red Hat EL 3.0 with enhancements, or their own thing.

      I doubt that Cray doesn't employ people who can tune a kernel...and quite well. The rest of it doesn't matter as much since we're talking HPC and not a normal server; likely no shell, none of the utilities you'd normally expect, and no GUI of any sort.

      I wouldn't be too stunned if the whole thing could fit on a floppy.

      --
      A firewall can not protect you from yourself. Turn off what you do not need. Do not use the firewall to do your work.
  25. Oh God... by Angst+Badger · · Score: 1

    Chassisses?

    Why do I have the sneaking suspicion that you're not pronouncing it correctly, either?

    --
    Proud member of the Weirdo-American community.
    1. Re:Oh God... by Anonymous Coward · · Score: 0

      Me thinksess his spellingsess is a little offsess

      Gollum.

  26. Re:Does the XD1 give the illusion of shared memory by IWannaBeAnAC · · Score: 1

    Hmm, I think now I was wrong; from reading further it looks like you do need MPI to get beyond 2-way. But there is enough memory bandwidth that single system image would probably work quite well.

  27. obsesive compulsive correction! by roadrunnerro · · Score: 1

    Actually this is AMD64 / x86-64 - it just emulates x86...

    1. Re:obsesive compulsive correction! by Anarke_Incarnate · · Score: 1

      no emulation, it supports x86 natively. You must have confused it for the EPIC chips from intel.

    2. Re:obsesive compulsive correction! by Waffle+Iron · · Score: 1
      All modern X86 processors emulate the instruction set. The internal hardware looks very little like the logical architecture view presented to the the programmer.

      The Itanium just happens to have a particularly poor X86 emulation implementation.

    3. Re:obsesive compulsive correction! by tesmako · · Score: 1
      One does not usually count microinstruction-based CPU's as having the microinstructions as the "real" instruction set, especially considering that one cannot actually get to the microinstructions of an x86 (on some older architectures the more complex instructions were microcoded in terms of simpler ones that were also usable, but it was still very much counted as part of the instruction set)

      The EPIC on the other hand does have a completely different instruction set with x86 emulation being a fairly separate core. I don't feel that their is any comparison.

      It is true though that strictly speaking all arguments about the x86 instruction set being something bad is kinda moot at this point, the exposed instruction set matters less and less with every day that passes. With the plentiful register renaming and out-of-order execution one does not even have to do an all that great job keeping the registers full and instructions well-scheduled to keep the CPU working with a decent level of efficiency.

      So, in summary, I feel that the x86 instruction set should be pretty much seen as a historical abstraction, we won't die from having it kicking around, the instruction decoding is a tad expensive but with the proper optimizations and cahcing in place it is quite livable. Beyond that the instruction set does not matter to a great extent in this day and age. It could also be noted that perhaps EPIC's greatest problem is that the more typical ad-hoc multi-issue OoO CPU has gotten so much better at its work that expecting to win all that much by betting on the compiler doing an excellent job at generating explicitly parallell instructions is looking like a bad idea considering that it locks the way the hardware handles things down completely (whereas the ad-hoc multi-issue OoO is completely abstracted).

  28. Bah. Crey. by Asmor · · Score: 1

    Every time I shut down one of their nefarious schemes, the Countess herself personally thanks me for rooting out corruption in her corporation. I'm beginning to think Crey's not as innocent as they claim... Wait, what? Oh, sorry, wrong Crey.

  29. Necessary by Dacmot · · Score: 0, Redundant

    Imagine a beowulf cluster of these...

  30. Still waiting for... by koehn · · Score: 5, Funny
    I'm still waiting for Cray to ship their pen-based system...

    ... wait for it ...

    Crayola!

    1. Re:Still waiting for... by mapmaker · · Score: 2, Funny

      If they used Opteron processors in it I'd imagine they'd call it the Crayon. No trademark issues that way either...

    2. Re:Still waiting for... by Fnord · · Score: 2, Interesting

      You think its a joke, but I used to work there and they do refer to employees of the company as Crayons.

  31. Will it run SAS? by Anonymous Coward · · Score: 0

    Sure, good benchmarks, but can I run SAS regression analysis on it?

  32. BS detector by spinlocked · · Score: 0, Troll

    Feel free to use my hardware company marketing BS detector:

    #!/bin/sh
    # BS detector
    grep "high bandwidth, low latency"

    Works every time, or your money back.

    --
    # init 5
    Connection closed.


    Oh... ...bugger.
  33. The should have renamed it... by rmy1 · · Score: 0

    ...it SKYnet.

    A rack or two of those, some AI software, and away we go!

  34. Hell, yeah! by DarkMan · · Score: 4, Interesting

    For my apps, I do iterative matrix calculations. However, one of the required data tables scales as n^2.3 (ish) of the system size. These can be precalculated, or calculated on demand. Typical size for a small run is 4-6 GB. I've filled a 40 GB array with data tables before.

    Thus, the part that impacts runtimes the most is either the on disc lookup, which is still faster than direct calculation, which we've also had to do.

    I looked into FPGA's a while back. Some back of envelope calculations show that a single FPGA should be able to calculated the data table on demand, and it'll be faster than reading from disc.

    (Turns out, that to actually get a usable solution for a basic PC would need to hack up the whole tool chain. FPGA cards for a PC are all designed for DSP, rather than numerics).

    So, with an FPGA and a CPU, I could elminated the slowest part of the job, and scale up to, what, a 1GB working matrix, which is about 8 time larger than the biggest job I've ever run, which hogged a T3E1200 for 6 hours.

    So, in short, gimme an FPGA and some reasonable tool chain, and I will be able to about half runtimes, and, more importantly, scale up to 10 times larger calculations. 5 time larger calculations is the most I've ever been asked about.

    Time to brush up on my VHDL, I think.

  35. XD1 == OctigaBay by Anonymous Coward · · Score: 1, Informative

    Cray did not develop this system themselves, they simply bought this little startup company and relabled its product.

    1. Re:XD1 == OctigaBay by telemonster · · Score: 1

      This is similiar I believe to how Cray "developed" the air-cooled J90 series.

      --
      Southeastern Virginia REPRESENT!
    2. Re:XD1 == OctigaBay by fgodfrey · · Score: 1

      Well, sort of. The J90 was a Cray design almost entirely. It came from Chippewa Falls (the historic design place of all Cray hardware). Its predecessor (the Y-MP-EL series) came from a startup company that Cray bought several years before J90 was done. This system was, for all practical purposes, done when we bought OctigaBay (now called Cray Canada) in April.

      --
      Go Badgers! -- #include "std/disclaimer.h"
  36. right.. but what happened to Tera?! by bmajik · · Score: 3, Informative

    i was looking at cray.com and there's no mention of the Tera MTA. The Tera MTA was the innovative idea they had to have 128 logical threads on a single CPU.. think of hyperthreading but with 128 logical threads instead of.. 2.. and also it was working at least 8 years ago.

    If you look at cray.com today its pretty sad. 3 product lines - the TD1 opteron+magic, the X1, which is traditional cray vector (smp vector nodes, and MPP's of those nodes), and their 3rd product line is the NEC SX-6... they're reselling it in the states for NEC.

    If you hit tera.com, you get a 404 :/

    --
    My opinions are my own, and do not necessarily represent those of my employer.
    1. Re:right.. but what happened to Tera?! by fitten · · Score: 2, Interesting

      Tera's MTA isn't exactly like HyperThreading. HyperThreading looks at (currently two) threads and sees which instructions from each stream it can schedule each clock. MTA was more like round-robin scheduling of threads on a per-clock basis. At each clock N, it scheduled an instruction from thread N, on clock N+1, it scheduled an instruction from thread N+1. In other words, if your process had only one thread, and the MTA processor ran at 1GHz and had 128 "threads", then your process ran as if it were on a single threaded 8MHz CPU.

    2. Re:right.. but what happened to Tera?! by Fnord · · Score: 2, Informative

      Well, partially. The scheduling did still have some severe benefits. Most importantly was for memory bandwidth. One of the threads can do a fetch, and the scheduler continues onward executing other instructions. As long as you can guarantee that a fetch doesn't take longer than 128 clock cycles, you've lost nothing. Which also means you can put a highly latent, extremely high bandwidth memory bus on there and reasonably expect to get 99.9% utilization of it. Combine that with some specialized instructions for working with big data sets and you have a pretty decent machine.

      Only problem was they couldn't get the damn thing to work for so long that it was practically obselete by the time they got it out. They sold all of 2 of them.

    3. Re:right.. but what happened to Tera?! by fitten · · Score: 1

      Yup... I saw one (well, at least the case of it... there was speculation if there was anything inside it) at SuperComputing. IIRC, it looked kind of like a giant piece of cake that a giant sat down on one side of it.

  37. Coincidence? by JediTrainer · · Score: 4, Funny

    Id releases Doom 3 for Linux, Cray announces availability of new supercomputer.

    Dare we say, we've finally actually found the hardware that can run this game?

    --

    You can accomplish anything you set your mind to. The impossible just takes a little longer.
    1. Re:Coincidence? by Scrab · · Score: 1

      i dunno. Might need a bit of overclocking....

      Just in case.....

      Scrab

      --
      RoseColor red={0, 0xffff, 0x0000, 0x0000};VioletColour blue={0, 0x0000, 0x0000, 0xffff};find / -name *mybase*|chown you
    2. Re:Coincidence? by shfted! · · Score: 1

      We even have a way to power it now. Coincidence? I think not!

      --
      He who laughs last is stuck in a time dilation bubble.
  38. Find the Fake Ad? Get a Life! by Saeed+al-Sahaf · · Score: 0, Flamebait
    These press releases ripped right off of PR newswire and posted to Slashdot always remind me of the "find the fake ad" section of Games magazine. "Find the fake news story planted on Slashdot to generate `buzz'". Bah. I hate marketing.

    So you're saying that /. is marketing for Crey? Have you considered that maybe the press release says what needs to be said? Have you also considered that most /. stories are summaries of complete stories? Very few (with the exception of reviews and "ask /.") Slash stories are "original material". That's what this site is: A NEWS DIGEST!

    Now then. The real question is: will this thing fit on my desktop?

    --
    "Who are in control, they are not in control of anything - they don't even control themselves!" - Glen Beck
    1. Re:Find the Fake Ad? Get a Life! by JackBuckley · · Score: 1
      Um, no, Slashbot. I am not saying that "/. is marketing for Crey [sic]." I am saying that much of what passes for journalism is actually driven by the marketing departments of corporations, who spoon-feed their press releases to media sources too understaffed or lazy to fill their own pages or broadcast. I am arguing that Slashdot should continue to be a "NEWS DIGEST!" not a "PR DIGEST!"

      And I have never seen a press release that "says what needs to be said." Press releases say what companies want to be said.

  39. In other news... by glMatrixMode · · Score: 3, Funny

    ... the Gentoo Chief Marketing Officer made the following statement :

    "We welcome the Cray XD1 as the first platform on which Gentoo installs in less than 12 hours. Looking forward to renaming Gentoo to 'One-Click-Linux'. Stay tuned !"

    --
    War doesn't prove who's right, just who's left.
  40. A good use for a bunch of these systems by ShatteredDream · · Score: 1

    Have the NSF buy a few billion dollars worth of high end Opteron hardware and make it availible to those who are doing public research. I think that access to a few million dollars worth of high end hardware would help cut down on the R&D costs for drugs that are partially paid for with taxpayers' dollars. The good side is that if part of the package is free time on really sophisticated NSF clusters capable of really cutting down number crunching time, the public can demand on its end lower prices and/or shorter patent intervals.

    1. Re:A good use for a bunch of these systems by Seanasy · · Score: 1

      The NSF just invested in a different Cray. However, your dream of drugs prices going down because of drug design on publically funded computers is a bit out there. The drug companies already take advantage of publically funded research to increase their profits. Besides, I think most of the biomed work on NSF supercomputers is much more basic than actually designing of the drug -- think protein folding and membrane channels. Yeah, it's important for drug design but far from actual production of a drug.

      So, it's already being done and, no, it won't reduce drug prices.

  41. Hmmm... by jd · · Score: 0
    Cray... Cray... yeah, that's the company that slagged off Linux clustering and claimed that vector processors were the way to go.


    Either the current management has a very different idea of what a vector processor is than did their founder, Seymour Cray, or they've become a bunch of corrupt, politiking schemers, no different from the Microsofts and Suns of the world.


    Either way, this actually marks a very sad day in Cray's history. Once the absolute bleeding-edge of technology, with each processor hand-designed and crafted at the transistor-level by Seymour himself, using the limits of cooling technology to keep the thing running.


    This machine is really not much different to SGI's Altix, except running the AMD processors rather than Intel. This means that although each processor likely runs faster than the ones SGI uses, Cray can't bundle as many together, as AMD hasn't progressed nearly as far on SMP-aware chipsets as Intel. (It's about the only area left where Intel has any kind of lead, which makes Intel's abandonment of the Xeon processors surprising. You'd have thought they'd try to capitalize as much as possible on what lead they do have left.)


    Now, I'm not opposed to Cray changing their minds, deciding that clustering is actually not a bad way to go, and THEN marketting a cluster-based machine. Hey, everyone's entitled to make mistakes, and it's great when people FIX their mistakes. As it stands, though, it merely looks like Cray is copying the success of Linux clusters and then trying to disparage Linux so that people will buy Cray's machines instead.


    Hey, build on success. Stand on the shoulders of Giants. But, whilst you're standing on them, try not to kick them in the head.

    --
    It's a small world and it smells funny; I'd buy another if it wasn't for the money; Take back what I paid (SoM)
    1. Re:Hmmm... by afidel · · Score: 4, Informative

      This machine is really not much different to SGI's Altix, except running the AMD processors rather than Intel. This means that although each processor likely runs faster than the ones SGI uses, Cray can't bundle as many together, as AMD hasn't progressed nearly as far on SMP-aware chipsets as Intel.

      This is some of the stupidest piles of drivel I have read on slashdot. SGI and Cray both do ALL of the glue logic chips themselves, that's the whole point of buying from them. They don't use the off the shelf chipset, they design their own with the design goal of large scalable systems. Besides Intel uses a shared bus where AMD uses the point to point bus they bought from Compaq which was origionally designed for the Alpha. So if anyone has a scalability lead it's AMD.

      --
      There are 4 boxes to use in the defense of liberty: soap, ballot, jury, ammo. Use in that order. Starting now.
    2. Re:Hmmm... by flaming-opus · · Score: 4, Informative

      Cray now has three product lines to address 3 different market segments.

      They have the X1, which is a massively parallel vector system for the very high-end. (For those who need 30+Gbytes/second of memory bandwidth for EACH cpu) These things are huge, expensive, and used by a limited number of users, mostly governments.

      They are getting ready to productize red storm, which is also a bunch of opterons, but strung together in a shared-memory system like the T3E. also a high-end solution.

      This system, the Xd1, is a low end system designed to be a half-step better than a cluster of off-the-shelf opterons. It's a multi-kernel cluster using MPI for all the data sharing. However the interconnect basically sits where the south-bridge sits on most opteron boxes.

      So Cray still has the absolute cutting edge systems, but have now expanded down-market. (Rather, they acquired octiga-bay who did the early design work).

      This is also not the first time this has happened. In the early 90s, Cray purchased a small start-up that was developing a NUMA-style mini-super based on sparc processors. They turned it into a product and sold a few, though not as many as they would have liked. During the SGI acquisition they sold the product to SUN, who branded it the E10000, and made about a billion dollars off of it. It's now the foundation for all of Sun's high-end Unix servers.

      Cray also bought a small company (I forget the name) that made a cmos implementation of the YMP. This became the ymp-el, the J90, which pioneered technology for the SV1.

      Cray has often built mid-range systems. Nothing new.

    3. Re:Hmmm... by flaming-opus · · Score: 1

      Most itanium systems use a shared bus.

      One should note, however, that the altix does not use the shared bus features of the itanium. Or at least that that bus is only shared by one cpu, the memory, and the bridge chip. The interconnect architecture of the altix is identical to the interconnect used on the old SGI origin systems, which were based on MIPS processors. From an architecture point of view, the Altix and the XD1 are very very similar. One uses itanium, one uses opteron.

      Altix tries to run a single OS image across the entire machine, while XD1 relies on MPI to do data sharing. However, even SGI doesn't spread a single linux image across their biggest machines. They also create a cluster-in-a-box on large enough configurations.

    4. Re:Hmmm... by fingusernames · · Score: 1

      This is also not the first time this has happened. In the early 90s, Cray purchased a small start-up that was developing a NUMA-style mini-super based on sparc processors. They turned it into a product and sold a few, though not as many as they would have liked.

      The SuperDragon! I worked at Cray Research in that timeframe, I have some SuperDragon memorabilia. When I was there, the T3D was under development, and was considered a mid-range supercomputer, beneath the C90, above the "entry-level" EL series.

      Larry

  42. Finally!! Now I can run DOOM 3 with FULL DETAIL! by Anonymous Coward · · Score: 0

    This is one spec that WILL not fail me!

  43. Hmmm.... by Dracolytch · · Score: 2, Funny

    I guess the chickens win after all.

    ~D

    --
    This sig has been enciphered with a one-time pad. It could say almost anything.
  44. Cray linux by floydman · · Score: 1

    Thats what it says on its specs sheet "Cray HPC-enhanced Linux, Kernel version 2.4.21"

    I think its wise they went with the 2.4 kernel though, but i wonder what is this cray linux, never heared of it.

    --
    The lunatic is in my head
    1. Re:Cray linux by Anonymous Coward · · Score: 0

      but i wonder what is this cray linux, never heared of it.

      Well, just as a wild guess, I'd say it's a customised Linux kernel/distro for Cray?

  45. Emoticons by Solder+Fumes · · Score: 1

    XD1 would probably be a good emoticon for anyone who manages to get their hands on one of these babies.

    1. Re:Emoticons by Lucretian · · Score: 1

      well then... XD1 to you! We got one at my place of employment a few weeks ago. It is a nice machine. We've been following it since it was an OctigaBay 12k system at SC03. It's basically the same system as it was then I guess, but with the Cray name as far as I can tell. Only a few complaints so far is they could give you a little longer power cord. The thing that comes with it is like 12". Also the case is a bitch to get into... it's like 12 screws, thumb screws would be a plus, something like the IBM p630s with a slide on cover or something would be a lot better. Otherwise the interconnect performance is great!

  46. IBM BlueGene by a3217055 · · Score: 1

    Firstly, Let's see 12 processors in a rack. That sounds great, but what happens when the chip blows... or something breaks how easy is it to fix. 58 GFLOPs ( I personally think it should be GFLOPS) . Anyway IBM's BlueGene will make Cray look like a digital watch. With already great performance we are gonna see a lot of changes we do computing. BlueGene is the future Cray is the clock telling us how far _into_ the future we are. Also another thing thank God ( please substitute ) for a 12-way x86 processor ( man what a waste ...). Hopefully this will cause better MPI programs to be written and become more main stream.
    Go Red Coats

  47. Pah! by mollymoo · · Score: 3, Funny

    Loads of Opterons? Who cares if GFX card is teh sux? Cray are a bunch of noobs. I bet it doesn't even have neon fans! You'll never get the chix showing them your 1337 skillz in CS with that heap of junk.

    --
    Chernobyl 'not a wildlife haven' - BBC News
  48. It runs open source codes! by Kurt+Gray · · Score: 1

    Quote the marketing info page: "runs wide variety of ISV applications and open source codes"

    That's the power of Cray's parallel processing: each machine runs its own "open source code" therefore a cluster is more powerful because the entire cluster runs "open source codes". At least that's their sales reps understanding of it.

  49. If you can't afford this Cray... by IronChefMorimoto · · Score: 3, Informative

    If you can't afford this Cray, you can at least buy the parts to start putting together your own multi-processor Opteron system:

    http://www.monarchcomputer.com/

    A friend of mine and I were talking the other night about local Atlanta, GA computer stores, and he mentioned that Monarch Computer is one of the only vendors from whom you can purchase the 4-way Opteron 800 series processors ($1200 a piece -- damn!).

    He's been in grad school out of state for a few years and was suprised to learn that Monarch Computer is, in fact, in his hometown backyard. Kind of kewl to walk in a store in your own town and walk out with a $1200 4-way processor.

    Until the wife finds out and sends you back to said store with the receipt in hand for a refund. :-\

    IronChefMorimoto

    P.S. - I don't work for these guys or advocate their store. I just thought it was cool to have such a vendor nearby. Too bad they don't sell Shuttle XPCs.

    1. Re:If you can't afford this Cray... by bobbozzo · · Score: 1

      Yes, but it won't be as fast for many applications unless you can get motherboards with hypertransport-infiniband bridges.

      --
      Nothing to see here; Move along.
    2. Re:If you can't afford this Cray... by Fnord · · Score: 1

      http://www.tyan.com/products/html/thunderk8qspro.h tml

    3. Re:If you can't afford this Cray... by bobbozzo · · Score: 1

      Thanks, but almost all Athlon64/Opteron boards have Hypertransport.

      What is needed is the Infiniband bridge.

      --
      Nothing to see here; Move along.
    4. Re:If you can't afford this Cray... by Anonymous Coward · · Score: 0

      "Kind of kewl to walk in a store in your own town and walk out with a $1200 4-way processor."

      What?!

      It's a 4-way capable chip and you only want to buy ONE!

    5. Re:If you can't afford this Cray... by lachlan76 · · Score: 1

      Ummmm.....aren't the 800-series opterons 8 way, 400-series 4 way, 200-series 2 way, and 100-series 1-way?

      I don't know server processors that well, but this is how I thought it works.

  50. Cray is giving customers what they want by einhverfr · · Score: 1

    They still make vector-based supercomputers, massively parallel systems, clusters, etc. These are designed to appeal to different markets.

    If you read Cray's 10-K, you will see that they believe that some computer problems can be resolved optimally on clusters, but others are better suited for vector-based systems.

    --

    LedgerSMB: Open source Accounting/ERP
  51. [sic] by Anonymous Coward · · Score: 0

    Does using that little "[sic]" thingy make you feel way smart? By the way, ever wonder how many "real" news stories started out as press PR? Quite a bit, actually.

    1. Re:[sic] by JackBuckley · · Score: 1
      Exactly my point, Anonymous Coward. Way too much of what passes for news in the United States starts out as VNR's or stories on PR Newswire. This is exactly what the marketing folks want--credible 3rd party reporting of a PR department's message. Justifying the value of PR by pointing to the number of PR stories that make it into a news cycle is circular and silly.

      And no, clarifying the source of a typographical error by using [sic] in a quotation does not make me feel "way smart." Being way smart makes me feel way smart.

  52. It's nice to see Cray out there by Proteus · · Score: 2, Insightful

    Cray systems may not always be the fastest thing around, but they are solid. It would be nice to see more producers paying careful attention to clean design and reliability over having the latest speed-booster.

    It's nice to see our old friend Cray continue to keep a foot in the market -- if nothing else, it makes everyone else stay on their toes.

    --
    We may not imagine how our lives could be more frustrating and complex—but Congress can. – Cullen Hightower
  53. We likes... by ZappaSoft · · Score: 1

    We likes chassisessess

  54. Whither MTA? by Troy+Baer · · Score: 2, Informative
    Look under initiatives on the Cray products page; the MTA-2 is shown there.

    The MTA idea is neat, but nobody's ever been able to find a problem that runs all that well on them. The original MTA didn't have enough memory bandwidth to make it competitve with a vector machine, and the small number of them in the field (less than 10 IIRC) are notoriously cantankerous. When Tera bought Cray, the one of the main things they were buying, aside from name recognition, was Cray's CMOS design experience; they were hoping Cray's designers could help with the problems they'd run into with MTA.

    --
    "My life's work has been to prompt others... and be forgotten." --Cyrano de Bergerac
  55. Bah... by Blue-Footed+Boobie · · Score: 1

    CRAY used to be so cool back in the day. Hell, the old CRAY systems are CHEAP now! http://cgi.ebay.com/ws/eBayISAPI.dll?ViewItem&cate gory=1484&item=5722664206&rd=1&ssPageName=WDVW

    --
    DAMN YOU OCTODOG! DAMN YOU TO HELL!
    1. Re:Bah... by Blue-Footed+Boobie · · Score: 1

      I really suck at the intarweb...

      --
      DAMN YOU OCTODOG! DAMN YOU TO HELL!
  56. Re:Does the XD1 give the illusion of shared memory by flaming-opus · · Score: 1

    NO.
    For that you need to buy crays mpp system "strider", which is a productized version of red-storm.

    The xd1 hardware is probably capable of shared memory, the software is not. The nodes (each 2-cpu blade) run off-the-shelf linux, and use MPI to share data.

  57. Let's See... is my G5 math right? by Wingsy · · Score: 2, Interesting

    12 Opterons deliver 58* GFlops (where * = peak). The Army's recent G5 cluster (1566*2 G5 processors running at 2GHz) deliver 25* TFlops. 58 divided by 12 yields 4.8* GFlops per chip for an Opteron, and 25000 divided by 3132 yields 8* GFlops per chip for the G5. What's wrong with this math? I didn't think the G5 had numbers THAT much better than an Opteron. And with G5s hitting 2.5GHz today the numbers would be much worse (or better, depending on your point of view).

    --
    If I didn't have absolutely NOTHING to do, I wouldn't be here.
    1. Re:Let's See... is my G5 math right? by RalphBNumbers · · Score: 1

      Your numbers are right.
      The Opteron's theoretical peak double precision floating point numbers are pretty mediocre, it gets beaten soundly by Xenons and Itaniums too. It's just not what the processor was designed for, it only has one FPU iirc, where the G5 and most of the other competition has 2.

      It is worth noting that these are theoretical numbers, not what you can actually achieve in reality on any given algorithm. The Opteron's Rmax is slightly more competitive(2.9GF/proc at 2Ghz in LosAlamos's Lightning), and the G5's Rmax (4.7GF/processor at 2Ghz in VT's machine) isn't quite as insanely high as it's Rpeak (8GF/proc).

      --
      "The worst tyrannies were the ones where a governance required its own logic on every embedded node." - Vernor Vinge
  58. What happened to RedStorm? by telemonster · · Score: 2, Interesting

    Cray not-too-long-ago had major announcements with the RedStorm project. I believe that system is supposed to be a single image 10,000 CPU AMD based rig. There are some oddities friends have pointed out, like the OS is based on IRIX I believe...

    Yea check this out:

    Cray Unicos/mp"

    Actually that references the X1, which is not based on PeeCee stuff, but actually a 8 core MPM.

    Sad thing is, even with Red Storm I think IBM will remain on top as their contract calls for 130,000 of their powerPCs on one system?

    It would be nice to see Cray on top, with something other than a commoditiy processors. I realize the T3D and T3E were both Alpha based systems.

    PS, I still have a J932se 32 proc Vector Cray ( for sale ) if anyone wants a Cray for home. $4500, real deal 3 cabinet Cray from 97', most likely used for gov't nuclear energy something-or-other. Located in Southeastern Virginia.

    --
    Southeastern Virginia REPRESENT!
    1. Re:What happened to RedStorm? by flaming-opus · · Score: 2, Informative

      It's still on-going. Red Storm is the focus of cray's effort over the next couple of years. Red Storm is the real-deal MPP-style system with a micro-kernel OS. xd1 is a low-end mini-super they acquired to expand down-market. (like mercedes buying chrystler)

      2 complimentary product lines. You could run the same application on both, though red storm provides real shared memory, which might allow better optimizations.

  59. Q404 by kcm · · Score: 1

    Sandia contracted out Red Storm development, hence RS is based on Sandia technologies towards petascale computing: Puma, Catamount, Portals, etc. No UNICOS, no IRIX, no Linux (except on the service nodes).

    This is a quintessential kernel architecture without support for threads, VM, IPC, etc. The interconnect is also asynchronous (5ns end-end speculated), point-to-point, and uses Portals.

    1. Re:Q404 by telemonster · · Score: 1

      Ahhhh that explains the shift in technologies from the X1 to RS/opteron! Much thanks for the infos.

      --
      Southeastern Virginia REPRESENT!
  60. Ooops... :-) by sczimme · · Score: 1


    I meant "696 GFLOPs per rack", not "per chassis", where a fully-loaded rack contains 12 chassis (58 * 12 = 696). D'oh! I appreciate the correction.

    --
    I want to drag this out as long as possible. Bring me my protractor.
  61. I love Cray by Superfreaker · · Score: 3, Interesting

    After searching everywhere for the legendary "Wang Computer" tshirt, I decided to fall abck on teh second geekiest computer company to get a shirt from, Cray. I couldn't find a shirt through the normal outlets (eBay/ThinkGeek), so I called them directly. The woman that answered was glad to help and shipped out, not a tshirt, but a very nice collared shirt that makes it look like I work for Cray! I wer it to all the conventions and I become cool(er).

    *queue calls to Cray*

    1. Re:I love Cray by spektr · · Score: 1

      a very nice collared shirt that makes it look like I work for Cray! I wer it to all the conventions and I become cool(er).

      /me sings "No woman, no Cray..." ;)

    2. Re:I love Cray by fingusernames · · Score: 2, Interesting

      Cray was always big on the memorabilia. I worked there as an intern for two years in the mid-90s. I still have many t-shirts (including our custom "World's Fastest Interns" version), some very cool posters, a ceramic Y-MP model, obligatory mugs, a cool 1-800-BUG-CRAY coaster, so on. I would call them up every so often after leaving there, and have them send me rolls of new posters and such.

      That was a unique experience... I had a security pass to the machine room, full of Cray C-90, Y-MP, X-MP, Cray-1 & Cray-2, lots of others. Awesome environment.

      Larry

  62. A Operton cluster??? by c0p0n · · Score: 1

    I would REALLY like to know what appereance a operton beowulf cluster would have...

    --

    Your head a splode
  63. Chassises? by Anonymous Coward · · Score: 0

    Is this a word?

  64. Oh, come on, did I miss it? by Anonymous Coward · · Score: 0


    Imagine a Beowulf Cluster of these!
    </obligatory>

  65. Dual-Core opterons by Guspaz · · Score: 1

    My math shows this to be a 6u unit (72/12=6)

    There are quad-opteron 1U boxes... So currently 6u of space can hold twice as many Opterons as these Cray units (24 Opterons with normal servers, 12 Opterons with the 6u Cray). The rapidly approaching introduction of dual-core Opterons would allow 48 opteron cores in the space this 12 opteron Cray.

    Yes, the Cray has many extras, (The FPGAs for example?), but for pure power, you might be better off with normal servers.

    1. Re:Dual-Core opterons by RalphBNumbers · · Score: 1

      More like a 3U unit (making it 36U for 12 servers).

      Since when are racks 72U high? It's more like 72 inches, or 42U, for the big ones.

      --
      "The worst tyrannies were the ones where a governance required its own logic on every embedded node." - Vernor Vinge
    2. Re:Dual-Core opterons by scoobrs · · Score: 1

      (Score:-1 Math-impaired)

      --
      -Those who would give up essential liberty to purchase temporary safety deserve neither. -Ben Franklin
    3. Re:Dual-Core opterons by Guspaz · · Score: 1

      My bad. In this case the Cray unit is just as dense as three quad opteron servers.

      However, that might change if dual-core opterons are introduced and Cray doesn't take advantage of them.

    4. Re:Dual-Core opterons by Anonymous Coward · · Score: 0

      Why in god's name wouldn't they?!?!?

  66. The end of custom CPUs by heroine · · Score: 2, Insightful

    Sort of sad they abandonned their custum CPUs for these commodity CPUs. Their liquid cooling was pretty nihilistic. You'd think there would be a lot to be gained from the old techniques of restricting everything to 64 bit operations, liquid evaporation cooling, and quad core parts.

    1. Re:The end of custom CPUs by rawgod0122 · · Score: 1

      They did not abandon their custom CPUS!

      They are making the X1 (no more sales because) they are moving to the X1e (1.5x the clock speed).

      The problem with these systems is the scaler unit. It runs at half the clock speed of the vector unit. So as soon as the code stops using the vector unit your screwed (scaler through put = 1/128 of the speed of the vector unit (64 length vector, half the clock speed)).

      The XD1 is their low end. They are doing this machine for two reasons.
      1. Clusters are eating them alive. Also witness the partnership will Dell.
      2. Did you notice the bit about commodity processors compined with special purpose? What they are looking at doing is combining, into the same machine (ie memory space) both AMD CPUs and their custom kit. This all of a sudden means they get a very good scaler unit with the best vector processor in the world. Being that they are in the same machine means you can get the best of both worlds.

      All of a sudden you have a machine that really kicks some arse.

    2. Re:The end of custom CPUs by flaming-opus · · Score: 1

      uh, close.
      The scalar unit of the X1 has a 400mhz clock and is dual-issue. (It's a MIPS derivative) It's a late-90s era risc chip. There are 4 scalar units on each MSP.

      The vector unit is clocked at 800mhz, and has 8 execution pipes per MSP. The vector unit does not compute the entire vector all at once, in fact you wouldn't want it to, as the pipelining of the vector unit is what helps mask memory latency in a vector-processor. The CPU dumps the contents of the vector registers into the 30+ stage pipeline 8 at a time, and can use the output of one operation as the input of another operation without commiting the value to a register (this is called chaining). Since the latency of this process is really large, it masks the latency of loading values from memory. (most vector machines are cacheless. The X1 has one, but earth simulator does not) However, the memory bandwidth and computational throughput are very high. This is why vector machines are GREAT for big memory problems, and TERRIBLE for code with lots of branching and short loops. (object oriented database code for example)

      Thus the theoretical peak performance of the scalar unit is half of the vector unit. However, most codes only use one of the scalar units, and that is used to direct the vector unit. Vector processors often realise 50% or greater of peak performance, though this requires a lot of application tuning. Most scalar machines are lucky to hit 10%, even with heavy optimisations. This has much more to do with the vector instruction set, and the HUGE memory bandwidth (16 channels of rdram) than with the theoretical throughput of the ALUs. ------ Vector processors are a lot less versatile than a scalar processor. They force the programmer to arrange the problem in a way that exposes a lot of parallelism, and to explicitely tell the processor about that parallelism. If the programmer is able to do that, and goes through the work, the processor exploits that organization, basically until it is limited by the memory bandwidth. A superscalar design will try to geuss about parallelism, but it's a difficult thing to do since the machine is so generic and versatile.

  67. Unfortunatly, no. by Penguinoflight · · Score: 1

    Id did a pretty lame job of porting Doom3 to linux, and hardly any interesting features are present. No alsa, no amd64... doesn't work with radeons.

    I wonder what John has been thinking... a few years ago he was on /. answering questions, pretty into the linux movement. Has he dumped us?

    --
    "And we have seen and do testify that the Father sent the Son to be the Savior of the World"
    1 John 4:14
  68. Re:Pakistan by Anonymous Coward · · Score: 0

    you were probably thinking to demonstrate some superior level of intelligence/knowledge than the parent poster, but you've instead demonstrated your ignorance.

    parent poster was alluding to the decades-old, unceasing competition between India and Pakistan ("competition" including several nasty wars, or rather, one long war - yet to, in fact, be actually over - punctuated by periods of ceasefires), and that now that India has it, Pakistan would be getting one too (a la nukes).

  69. Please stop, you are hurting me. by Anonymous Coward · · Score: 0

    Their liquid cooling was "pretty nihilistic"? You do realize that you can't just pick a word that you think sounds cool, and pretend it means cool, right? Its sad when you make the 80's "rad" kids sounds smart.

  70. Imagine ... by pgfault · · Score: 0, Redundant

    Could you imagine building a beowulf cluster of these?

    Oh... Nevermind.

  71. I don't know, but.. by raehl · · Score: 1

    If you get the Operton Prime, it can also drive itself around.

  72. Re:Does the XD1 give the illusion of shared memory by tygr007 · · Score: 1

    Cant one use some parts of NUMA kernel?

    It was designed by SGI after they've acquired Cray ...

    Well anyway, I suppose you can run something like http://www.mosix.org/ on the XD1 racks :)

  73. 12 chasses of 12 CPUs... by jonadab · · Score: 1

    > Each XD1 chassis has up to 12 AMD Operton processors. Up to 12 chassis
    > can be clustered together in a rack.

    Man, that's just ... (drumroll please...) gross.

    --
    Cut that out, or I will ship you to Norilsk in a box.
  74. Good Colors Now Available by Anonymous Coward · · Score: 0
  75. Chassises? by Bnonn · · Score: 1

    What do Slashdot's editors do again?

  76. Or a fountain by myklgrant · · Score: 1

    Or a fountain
    http://www.csm.ornl.gov/ssi-expo/cray2.jpg/
    The coolest computer ever.
    Michael

  77. Just ask Hank Dietz! by Anonymous Coward · · Score: 0