Forgot your password?
typodupeerror
Supercomputing Technology Build

Homebrew Cray-1 140

Posted by CmdrTaco
from the when-i-was-a-child dept.
egil writes "Chris Fenton built his own fully functional 1/10 scale Cray-1 supercomputer. True to the original, it includes the couch-seat, but is also binary compatible with the original. Instead of the power-hungry ECL technology, however, the scale model is built around a Xilinx Spartan-3E 1600 development board. All software is available if you want to build one for your own living room. The largest obstacle in the project is to find original software."
This discussion has been archived. No new comments can be posted.

Homebrew Cray-1

Comments Filter:
  • by Skyshadow (508) * on Tuesday August 31, 2010 @01:03PM (#33427028) Homepage

    Back when I was an intern with SGI, we took a day off* to visit the Chippewa Falls Museum**, which has a good-size collection of Control Data and Cray Supercomputer relics along with other items relevant to my interests***.

    I got to poke around inside of an original Cray-1. To me, the most interesting thing about it was just how insanely packed the internal wiring was; I'd been expecting the intricate plumbing, but the sheer volume of wires running from Point A to Point B was really impressive. I mentioned this to the guy giving the tour, who turned out to be a retired manufacturing supervisor -- he told me that the hardest part of his job was finding women with both enough skill and small enough hands to handle the internal wiring jobs. The thing had been assembled *by hand*, every connection in this crazy bulk of wired clipped or soldered into place one after another.

    Anyhow, after that I sat on the couch. It was not comfy.

    * My boss was *pissed* about this -- she went around telling anyone that would listen that "interns are here to work, not go sightseeing". This marks the one and only time in my career that anyone in HR has ever done anything worthwhile, calling her up and telling her this was part of the program and she didn't get a vote.
    ** Seymour Cray moved to Chippewa Falls, his hometown, when he was still with Control Data because he felt most middle managers wouldn't want to drive that far just to bother him. Visionary man, that Seymour Cray.
    *** Stuff from Leinenkugals.

    • by blair1q (305137) on Tuesday August 31, 2010 @01:16PM (#33427186) Journal

      IIRC, they also deliberately hired both tall and short women, and assigned the appropriate subset of connections to each.

    • Re: (Score:3, Funny)

      Anyhow, after that I sat on the couch. It was not comfy.

      +1 Informative

      • by fyngyrz (762201)

        I agree, I always wondered about that couch. :)

      • by Caerdwyn (829058)

        Anyhow, after that I sat on the couch. It was not comfy.

        I guess the measure of the comfort of integrated seating on supercomputers is a kind of "bench"mark...

    • Re: (Score:3, Interesting)

      by StayFrosty (1521445)
      I've lived within a 2 hour drive of Chippewa Falls for many years and still have not visited the museum. I should look it up next time I'm passing through. In case you are ever in the area again, the Leinenkugals brewery gives excellent tours (with samples) as well and is IMHO worth a visit.
      • by Tom Arneberg (93330) on Tuesday August 31, 2010 @03:32PM (#33428874)
        The originals are definitely worth seeing! I am surprised how few people take an hour to see the Cray museum (now called "Chippewa Falls Museum of Industry and Technology", http://cfmit.org/ [cfmit.org]). The museum used to be housed in the Engineering building (where I'm typing from right now). I had to get a special camera pass to take our family Christmas Card picture with a Cray 1 in 1991: http://arneberg.com/family/xmas/xmas_card.cgi?1991 [arneberg.com] (Sorry about the photo quality...those scans were made in the mid-90s; I really should re-scan with modern technology!)

        BTW, the Leinie's Lodge is also well worth visiting! It's less than a half mile from the museum, and is actually the number one tourist attraction in at least a 50-mile radius.

        • by Skyshadow (508) *

          I used to work in Engineering -- this takes me back. Who owns those building now? Celestica? The new Cray spin-off?

      • by dcraid (1021423) on Tuesday August 31, 2010 @08:01PM (#33431394)

        For those of you in/visiting the DC area you can check out a couple of old Crays @ the National Cryptologic Museum on the outskirts of Ft Meade. http://www.nsa.gov/about/cryptologic_heritage/museum/virtual_tour/index.shtml [nsa.gov]

        On exhibit in the museum are two Cray supercomputers. The XMP-24 on display is the upgrade to the original XMP-22 that was the first supercomputer Cray ever delivered to a customer site. It was in operation from 1983 to 1993 and was arguably the most powerful computer in the world when it was delivered. It used serial processing to conduct 420 million operations per second.

        The second generation Cray, the YMP, replaced the older version in 1993. It had a 32 gigabyte (32 billion bytes) memory capacity. In 1993 most personal computers held only 16 million bytes. The YMP used vector processing, a very powerful form of overlapping, parallel processing to conduct 2.67 billion operations per second. The YMP was decommissioned and went on display at the museum in 2000.

        The museum is lots of fun and definitely worth a visit.

    • by drfuchs (599179) on Tuesday August 31, 2010 @01:47PM (#33427576)
      "Why," you may ask, "was the internal wiring so insanely packed?" The length of each point-to-point wire was individually calibrated, such that all the signals to each gate arrived at the same moment, so you didn't need flip-flops to latch values in the flow of the circuits. Kind of a "just-in-time delivery" of electrons; and each layer of buffering avoided saved you delay along the pipeline. I don't think this sort of scheme was used on any other mainframe.
      • by conspirator57 (1123519) on Tuesday August 31, 2010 @04:11PM (#33429310)

        a shorter way of saying this is "phase matched cables".

        http://www.mwjournal.com/journal/article.asp?HH_ID=AR_7184&tite=Phase%20Matching%20and%20Tracking%20of%20Coaxial%20Cable%20Sets [mwjournal.com]
        pay no mind to the fact that it's discussing microwave applications. it's just as relevant to high speed digital (or lower speed, further distance) digital electronics. keep in mind how far this interconnect had to travel due to part density and cooling needs.

        of course for modern electronics one might use tunable LC networks on either end to accommodate variations in commodity sourced cables.

        also, i take issue with ECL being described as inefficient. CMOS is inefficient when transitioning a lot. ECL merely has a constant power drain, which also happens to make it less noisy. It was also differential before LVDS was cool (or even around).

      • "Why," you may ask, "was the internal wiring so insanely packed?" The length of each point-to-point wire was individually calibrated, such that all the signals to each gate arrived at the same moment, so you didn't need flip-flops to latch values in the flow of the circuits.

        Deskewing the signals didn't eliminate the need to flop the levels - by a long shot.

        What it's about is arranging the placement of gates and routing of signals so you avoid situations where most of the inputs to a logic function have arri

      • "Why," you may ask, "was the internal wiring so insanely packed?" The length of each point-to-point wire was individually calibrated, such that all the signals to each gate arrived [after a known amount of time delay] ...

        Cray did this from his earliest designs for Control Data.

        His first machines were the CDC 160 (a low-power I/O processor) followed by the CDC 1604 (his first "mainframe"). I "cut my hacker teeth" on a 1604-B, serial number 6.

        That was a 48-bit machine (with two instructions per word to save

    • In the lobby in front of the old supercomputer center. (The new supercomputers are up the road in Laramie where electricity is cheaper and the vice-residents home state at the time.)
      • by treeves (963993)

        You mean NCAR don't you?
        I visited there when my Grandma still lived in Boulder back in 1986, and saw the Cray which, IIRC, was still operational at that time.

      • NCAR Cray 1-A (Score:5, Interesting)

        by cruff (171569) on Tuesday August 31, 2010 @03:05PM (#33428568) Homepage

        The new computing center is still under construction, so no supercomputers are located there yet. :-) Cray 1-A serial number 3 is in the hallway as an exhibit on floor 1B, and if you are near the NCAR Mesa Lab, it is open for self guided tours most days of the year.

        Serial number 3 was in active use until its decommissioning in 1989 and ran COS (Cray Operating System). It was connected to the NCAR designed Hyperchannel network known initially as the NCAR Local Network (NLN) and later as the Mainframe and Server Network (MASnet). There were rows of 100 MB and 200 MB "washing machine" disk drives connected to it, and it had access to the NCAR Mass Storage System (MSS) for archival storage. Graphical output could be sent to plotters, large Xerox printers, B&W microfiche or B&W or color 35mm film. For a speed comparison, I once ran a Madelbrot generator to produce a 640x480 image on both the Cray 1-A and a Sun 60 workstation. The vectorized C code on the Cray took just under 8 seconds, the Sun several minutes to produce the image. Alas, we don't have much in the way of documentation anymore nor is there any of the original software.

        • by treeves (963993)

          Very interesting. Thanks for that. Sounds like you worked there.
          How did the direct to film imaging work? CRT though optics to film?

          • Re: (Score:3, Informative)

            by cruff (171569)

            Very interesting. Thanks for that. Sounds like you worked there.
            How did the direct to film imaging work? CRT though optics to film?

            Yes, I still do work there. The film output did indeed work like you suspect. They were made by Dicomed, and we ran them so hard that our in house electronics maintenance personnel were well acquainted with them. The Dicomeds supported a 4096x4096 vector graphic display, and the usable portion depended on the output format and frame size. Raster output could be performed by displaying each point for the required amount of time to produce the desired exposure. The color versions had a color wheel and re

    • I've got mod points, but you're on a +5 already and the time taken to type this feels more worthwhile somehow.

      Thank you. Thank you, thank you, thank you. I've always wondered how comfy those seats were, nearly as much as I've wondered why the hell they were there in the first place. Well now I know, and will die ever so slightly happier because of it. That's the most interesting thing I have ever read on Slashdot and I salute you for it. "It was not comfy". Thank you sir.

      (And I'm being utterly
      • Troll? No, really, I meant every word. Out of everything in IT the one thing that's always bugged me is the Cray couch. I've no idea why it's there, and frankly a review of its comfort is the best I've had in 20-odd years. I'm not posting to complain about the troll, hell, I once got a -26 that I'm kind of proud of (and still stand by), I just don't want the OP thinking I was taking the piss. Seriously, most interesting Slashdot comment ever. Please contact me via my sig if you're in any doubt.
      • ... nearly as much as I've wondered why the hell [the seats] were there in the first place.

        As I recall the Cray was curved with wedge-shaped boards and wiring on the inside of the curve to minimize the length of interconnecting wires and thus maximize logic speed. The outside of the tower was for access to the cards. And that meant the power supplies, cooling plumbing, and other support miscellany had to go somewhere else. "Somewhere else" was further out and at floor level - to be out of the way of acce

    • by nurb432 (527695)

      The wiring was the key to why it was so fast. All the interconnects. Something this *emulation* wont really have.

  • Xilinx... (Score:5, Informative)

    by TrisexualPuppy (976893) on Tuesday August 31, 2010 @01:05PM (#33427046)
    I built a PVP11 "supercluster" and started with Xilinx. The hardware is great, but their software toolset is horrendous.

    After months of free time development, I switched over to surplus Altera Stratix II video decoder hardware, got a copy of Quartus II, and was moving within weeks. Altera would be my suggestion for any geek who wants to try something similar!
  • by Anonymous Coward on Tuesday August 31, 2010 @01:07PM (#33427082)

    Not even a cray supercomputer can survive the slashdot effect.

  • Wow! (Score:5, Interesting)

    by line-bundle (235965) on Tuesday August 31, 2010 @01:08PM (#33427086) Homepage Journal

    Now that's closer to true News for Nerds, Stuff that Matters. Not the Mac fish tanks.

    However, I am a little disappointed that he didn't do something with the tower part of his cray. Cooling perhaps? Blinkenlights?

    How does it compare in performance with the original?

    • Re:Wow! (Score:5, Informative)

      by Neon Spiral Injector (21234) on Tuesday August 31, 2010 @01:13PM (#33427144)

      It's instructions execute accurately clock-for-clock, but running at 33 MHz instead of 80.

      • Re: (Score:3, Funny)

        by MarkRose (820682)

        Why, cray tell, does it run so slowly?

        • Re: (Score:3, Informative)

          by vlm (69642)

          Why, cray tell, does it run so slowly?

          I have one of those, the Spartan board, not a Cray-1. I did not remember, but checked online and the Spartan board has a 66 MHz canned oscillator. So, his design probably uses two clock cycles per instruction cycle.

          Probably also limited my the memory speed of whatever he's using for memory. 33 MHz equals what 30 ns access cycle?

          • Re:Wow! (Score:5, Informative)

            by Space cowboy (13680) on Tuesday August 31, 2010 @01:44PM (#33427558) Journal
            S3E's have DCMs (Digital Clock Managers) making them very flexible [xilinx.com] in terms of what the internal clock frequencies are, even with a fixed input frequency.

            Chances are (I can't get to the site) it just runs at 33MHz as its best-supported clock frequency. An S3E is a pretty cheap and slow FPGA - I remember writing a 32-bit CPU for one, and until I started optimising the logic-placement in the FPGA, it was only running at ~30MHz. I got it up to ~50MHz after tweaking and pipelining, but his design may do more than my simple CPU.

            Simon
            • The S3E itself can be clocked internally at 300+ MHz. However, the maximum speed achievable depends on the architecture and layout of the circuit implemented. The maximum clock is dependent on the longest logic and routing delay through the circuit. Since the design is apparently a register for register copy of the original Cray architecture, the original ECL logic still has a speed advantage over the CMOS S3E.

              • Re: (Score:3, Informative)

                by Space cowboy (13680)

                The S3E itself can be clocked internally at 300+ MHz. However, the maximum speed achievable depends on the architecture and layout of the circuit implemented. The maximum clock is dependent on the longest logic and routing delay through the circuit. Since the design is apparently a register for register copy of the original Cray architecture, the original ECL logic still has a speed advantage over the CMOS S3E.

                Not having seen the design, I don't know how it's been implemented, but it's possible to have a compatible design that implements all the original specifications without designing it the same way... It's also possible for an FPGA design to run faster than the original part - see the multiple-tens-of-MHz variants of the equally venerable 6502 (which maxed out at ~2MHz at the time) for example.

                Clearly, the logic path dictates the final speed. That's why placement is so important, and why hand-placement is

                • Yes, the 6502 is equally venerable. No, the Commodore 64 or Atari 600 XL was not the same class of machine as a Cray 1 supercomputer. The 6502 did not have vector registers nor twelve pipelines. It didn't handle 32 megabytes of memory. It was a microprocessor, which is a name it gets from being three or four classes of computer below a supercomputer.

                  supercomputer > mainframe > minicomputer > microcomputer workstation > home and hobby microcomputer

                  • Um, did I claim that the Cray-1 was in the same class as a 600XL ? I don't believe so. I was using an example of how a chip of that vintage has been implemented with speeds approaching 50x the original.

                    To address your points individually, however:
                    • A vector register is simply local storage that some operations know how to use well - the Cray 1 had 8 registers, each holding 64 64-bit locations, for a total of 4096 bits per register. That's a trivial thing to implement using BlockRam on an S3E. I'd frankly
                    • Yes, "a chip" of that vintage. A 2 Mhz MOS microprocessor. Yet the Cray wasn't a single microprocessor. Getting a 50x speedup of a single MOS microprocessor on a single newer FPGA chip is one thing. Getting a 50x speedup from a wire-and-lead NAND array machine running at 80 MHz on a single FPGA isn't quite the same thing.

                      Not only were the data items 64 bits, but the vector registers could each be fed a 64-bit word each clock cycle. The four 64-bit instruction buffers could be completely filled with the 16-b

                    • At this point, I'm guessing you're just a troll. I've claimed nothing that you're arguing against, you're just putting words into my mouth and then making the case against them. Good strawman argument there. End of discussion.

                      Simon.
                    • Not having seen the design, I don't know how it's been implemented, but it's possible to have a compatible design that implements all the original specifications without designing it the same way... It's also possible for an FPGA design to run faster than the original part - see the multiple-tens-of-MHz variants of the equally venerable 6502 (which maxed out at ~2MHz at the time) for example.

                      Um, did I claim that the Cray-1 was in the same class as a 600XL ? I don't believe so. I was using an example of how

        • Re:Wow! (Score:4, Interesting)

          by nacturation (646836) * <nacturation.gmail@com> on Tuesday August 31, 2010 @01:49PM (#33427600) Journal

          Why, cray tell, does it run so slowly?

          You do realize the Cray-1 is from the late 1970s, right? 80MHz with this architecture was smoking fast.

      • Re:Wow! (Score:4, Insightful)

        by firewood (41230) on Tuesday August 31, 2010 @01:39PM (#33427490)

        Half the fun of explaining the Cray 1 during museum tours was comparing its cycle time to the time it took light from the nearby ceiling spotlight to hit the Cray. At 33 MHz that would require a really tall room.

  • by blair1q (305137) on Tuesday August 31, 2010 @01:12PM (#33427142) Journal

    because it just got slashdotted...

  • No software (Score:5, Interesting)

    by dk90406 (797452) on Tuesday August 31, 2010 @01:13PM (#33427152)
    He has the needed software for the FPGA, but he has (so far) been unable to find some software to run on the machine. At least that is what I got from the TFA. It seems like no-one (including various 3 letter agencies) have copies of stuff so "old".
    Never the less, I have to admire the effort put into this.
    • Re:No software (Score:5, Interesting)

      by permit594 (677033) on Tuesday August 31, 2010 @01:30PM (#33427382)
      The original Crays were delivered with no software -- not even an OS. We had to roll our own OS. I started at Sandia Labs in 82 as a fresh PhD. I still had some work to do on my software-based dissertation, so I got to play on the "new" Cray. I had been used to waiting a *long* time for my (FORTRAN) program to compile on UCLA's IBM 360... The first time I compiled on the Cray, it finished so quickly I thought I had a syntax error in my job submission command. For the times, that machine was FAST!
    • by WrongSizeGlass (838941) on Tuesday August 31, 2010 @01:35PM (#33427438)

      He has the needed software for the FPGA, but he has (so far) been unable to find some software to run on the machine. At least that is what I got from the TFA. It seems like no-one (including various 3 letter agencies) have copies of stuff so "old".

      Maybe he'd have better luck finding old software if ageism wasn't rampant in the IT industry [slashdot.org] ;-)

  • Pretty cool! (Score:5, Insightful)

    by oldhack (1037484) on Tuesday August 31, 2010 @01:15PM (#33427168)

    This is the sorta hack that should feature on the front page, instead of machining tin can case and similar tripes.

    Hope he gets some software for that thing.

  • But (Score:2, Funny)

    by SnarfQuest (469614)

    What version of Windows does it run? I might have a copy here...

    Doesn't everything run Windows?

    • Re:But (Score:4, Informative)

      by rubycodez (864176) on Tuesday August 31, 2010 @01:31PM (#33427394)

      well, this is a mid 70s computer, so it must have run CP/M 8D

      if he really does want to run real Cray software, he'll have to implement the interrupts and context switching for Cray Operating System (COS) or the Unix Unicos

    • Re:But (Score:5, Insightful)

      by WrongSizeGlass (838941) on Tuesday August 31, 2010 @01:32PM (#33427398)

      Doesn't everything run Windows?

      Everything runs Linux, but Windows seems to run everything.

      • Re: (Score:2, Funny)

        by kamochan (883582)

        You misspelled ruin.

      • by guruevi (827432)

        Sad, but true. I sometimes get scared when I see what runs on a networked Windows (XP). Neonatal monitors, respiration controllers, fMRI consoles, oscilloscopes, websites...
        The problem with most of these things is that they never can get updated to the latest service packs (and thus all run without Service Pack or Service Pack 1) or have any other software (like virus scanners) installed because that would void the warranty/FDA approval/application support.

      • by slick7 (1703596)

        Doesn't everything run Windows?

        Everything runs Linux, but Windows seems to run everything.

        At NASA Lewis/Glenn there was a Cray XMP that was later upgraded to a YMP. To run programs on the Cray required that the program be compiled on an IBM 360. There were two IBM's there. I was involved with the installation of power cables to peripheral equipment and modifications to the 400Hz UPS.

    • Turns out the system requirements aren't as stringent as once thought.
  • Where's the VAX? (Score:3, Interesting)

    by sconeu (64226) on Tuesday August 31, 2010 @01:30PM (#33427388) Homepage Journal

    When's he going to build the VAX front end? TFA alludes to the fact that the Cray-1 needed a dedicated mini as a front end, and sounds like he might actually get a diskpack from one (or image thereof).

  • theoretically, ECL would trounce any CMOS tech fabricated with current accuracy. It's just that it's so horribly power-hungry, that nobody will do this. For one thing, it would need megagalactic cooling.

    • Although individual gates of ECL could run faster than individual gates of CMOS or even dynamic NMOS, ICs of useful complexity for single-chip modern CPUs are not likely to be competitive if implemented in ECL. The problem is power density. Modern fast CPUs are already dissipation limited. ECL is inherently a 5 volt technology (I suppose cleverness could improve that somewhat) so it starts with a 4:1 heat disadvantage compared to MOS, and it only gets worse because each gate must always have current flowing

    • theoretically, ECL would trounce any CMOS tech fabricated with current accuracy. It's just that it's so horribly power-hungry, that nobody will do this.

      CMOS beats ECL in integrated circuits because you can put a LOT of gates VERY CLOSE TOGETHER without cooking them. This means that, though you lose on gate speed, you gain more back in short propagation time on the interconnects.

      ECL is inherently hot because each gate starts with an analog current regulator, followed by logic that "steers" the current into

  • NCAR (Score:5, Informative)

    by Fishbulb (32296) on Tuesday August 31, 2010 @01:56PM (#33427714)

    Send an email to the folks at the CISL [ucar.edu] division of NCAR [ucar.edu].

    They know a thing or two [ucar.edu] about Crays.

    • Re:NCAR (Score:4, Informative)

      by cruff (171569) on Tuesday August 31, 2010 @03:09PM (#33428598) Homepage

      See my reply earlier on in this discussion. We didn't keep any of the Cray software, unfortunately.

      • by Nyder (754090)

        See my reply earlier on in this discussion. We didn't keep any of the Cray software, unfortunately.

        do you regret that now?

        • by cruff (171569)

          do you regret that now?

          I do, but unfortunately I wasn't part of the group that managed the Crays, and I didn't think to track down the documentation and distribution media to archive copies to more modern media.

  • Really cool (Score:5, Insightful)

    by cygnwolf (601176) on Tuesday August 31, 2010 @02:00PM (#33427760)
    Again one of those instances where it'd be nice to be able to mod articles. This is the kind of stuff that needs to be on slashdot.
  • by ArcherB (796902) on Tuesday August 31, 2010 @02:15PM (#33427918) Journal

    From TFS:

    All software is available [chrisfenton.com] if you want to build one for your own living room. The largest obstacle in the project is to find original software."

    Um... why not just click on the little link provided there?

    • by Jeremy Erwin (2054) on Tuesday August 31, 2010 @02:39PM (#33428206) Journal

      It's obvious from your comment that you haven't downloaded and inspected his source code. It includes some verilog files for making the FPGA behave like a Cray-1, and some python files for debugging it and loading opcodes into the simulated cray. However, if you want to run vintage 1970s computer applications----weather simulations, cryptanalysis, computational fluid dynamics, etc., you would be hard pressed to find any.

  • by garyebickford (222422) <gar37bic.gmail@com> on Tuesday August 31, 2010 @02:22PM (#33428008)

    I think there were (are?) four of Supercomputer Centers that had Cray 1 and later Cray X-MP machines. The Pittsburgh center did a lot of work with Carnegie Mellon, esp. the Robotics Institute.

    I personally did one bit of work - porting a photometrically correct ray-tracer by Dr. Robert Thibadeau in the Image Understanding Laboratory from an Apollo workstation to the Cray at PSC - this would have been in 1989, I think. The one complication we had was that the Cray floating point format was different, so our first runs were all zeros. Other than that the code compiled and ran fine on the Cray. Of course, a run that took two weeks on the Apollo ran in about 40 seconds on the Cray.

    A lot, maybe all of the work done on these machines was non-spooky research so perhaps you can track some of the professors at the associated universities, such as CMU, Northern Illininois, UCSD, Berkeley, etc. Also check out the weather folks - they have been among the biggest CPU cycle-burners for a long time. I worked briefly with one weather guy at a weather research facility in Wyoming but I don't recall any details - was it U Wy?
    The SCs I recall are:

    • SDSC (San Diego Supercomputer Center),
    • PSC (Pittsburgh Supercomputer Center).
    • NCSA (National Center for Supercomputing Applications)

    I'm sure that if you dig around in the universities you'll find folks who have stuff piled on a back shelf somewhere (probably in a tape format you can't read). Also look up in the old annals of the ACM SIG on supercomputing - that will give a line on researchers who were working on the Cray.

    • Re: (Score:3, Funny)

      by LWATCDR (28044)

      You mean that nobody has ported GCC to it yet? It must run NetBSD right?

      Shocked I tell you just shocked.

  • by khb (266593)

    If memory serves (and wikipedia can help for those who don't recall), the first systems were shipped with just an assembler. Livermore and Los Alamos pulled together the first OS for it. COS (the Cray Operating System) came a little later. The Labs stuck with their own OS for quite some time.

    I haven't seen a source bundle, but since the OS shouldn't have been classified and it was paid for 100% tax dollars, someone probably could score a copy (Freedom of information request??).

  • ARTICLE TEXT (Score:5, Informative)

    by Brietech (668850) on Tuesday August 31, 2010 @02:38PM (#33428198)

    As part two (see previous attempt) of my ongoing series in ‘computational necromancy,’ I’ve spent the last year and a half or so constructing my own 1/10-scale, binary-compatible, cycle-accurate Cray-1. This project falls purely into the “because I can!” category - I was poking around the internet one day looking for a Cray emulator and came up dry, so I decided to do something about it. Luckily, the Cray-1 hardware reference manual turned out to be useful enough that implementing most of this was pretty straightforward. The Cray-1 is one of those iconic machines that just makes you say “Now that’s a super computer!” Sure, your iPhone is 10X faster, and it’s completely useless to own one, but admit it . . you really want one, don’t you?

    The Cray-1A Architecture

    Now, let’s get down to specs - What is this bad boy running? The original machine ran at a blistering 80 MHz, and could use from 256-4096 kilowords (32 megabytes!) of memory. It has 12 independent, fully-pipelined execution units, and with the help of clever programming, can peak at 3 floating-point operations per cycle. Here’s a diagram of the overall architecture:

    cray_architecture

    It’s a fairly RISC-y design, with 8 64-bit scalar (S) registers , 8 64-bit/64-word vector (V) registers, and 8 24-bit address (A) registers. Rather than a traditional cache, it uses a ’software-managed’ cache with an additional 64 64-bit words (T registers) and 64 24-bit words (B registers). There are instructions to transfer data between memory and registers, and then register-to-register ‘compute’ instructions.

    One of the coolest aspects of this machine is that everything is fully pipelined. This machine was designed to be fast, so if you’re careful, you can actually get one (or more) instruction every cycle. This has some interesting implications - there’s no ‘divide’ instruction, for instance, because it can take a variable amount of time to finish. To perform a divide, you need to first compute the ‘reciprocal approximation’ (something we *can* do in exactly 13 cycles, it turns out) of the denominator value, and then perform a separate multiply of that result with the numerator.

    The vector instructions are particularly cool. A vector Add operation might take only 5 cycles to start producing results (remember, each vector can hold 64 values, so it takes 5 + 64 cycles to finish adding). Why wait for it to finish though? We can take the result output from the adder, and “chain” it straight into another vector unit (say a multiplier). And *that* only takes another 10 cycles or so, so we can chain that result into yet another unit (say, reciprocal approximation). Now, rather than waiting for the first operation to finish, we’re computing up to 3 floating point calculations per cycle. Clever programmers could sustain about 2 floating point operations per cycle, or 160 million instructions per second.

    vector_chainingVector Chaining in Action!
    The Hardware

    The actual design was implemented in a Xilinx Spartan-3E 1600 development board. This is basically the biggest FPGA you can buy that doesn’t cost thousands of dollars for a devkit. The Cray occupies about 75% of the logic resources, and all of the block RAM.

    spartan3_1600

    This gives us a spiffy Cray-1A running at about 33 MHz, with about 4 kilowords of RAM. The only features currently missing are:

    -Interrupts

    -Exchange Packages (this is how the Cray does ‘context-switching’ - it was intended as a batch-processing machine)

    -I/O Channels (I just memory-mapped the UART I added to it).

    If I ever find some software for this thing (or just get bored), I’ll probably go ahead and add the missing features. For now, though, everything else works sufficiently well to execute small test programs and such.
    The Software

    When I started building this, I thought “Oh, I

  • by sootman (158191) on Tuesday August 31, 2010 @03:21PM (#33428736) Homepage Journal

    If you were plowing a field, which would you rather use? Two strong oxen or 1024 chickens?
     
    Seymour Cray [google.com]

    • How much to cool the chickens and coordinate the yokes?

      Since this is what many clusters do -- HPC that used to be on custom-built supers -- the quote is still as cool as ever but the question is now far from rhetorical.

  • the REAL news story here is that the guy is actually married!!

    damn... my wife doesn't know what a Cray is, doesn't care what a Cray is and certainly wouldn't want to help me paint it....
    in other news, i approve of the Bailey's and Stoli bottles on the table behind the mini-cray...
  • wow (Score:2, Funny)

    by Vorpix (60341)

    imagine a beowulf cluster of these! ;)

  • by braindrainbahrain (874202) on Tuesday August 31, 2010 @03:48PM (#33429062)

    Ages ago, I heard this story. Can anyone confirm if this is true or not?

    Seems Steve Jobs, upon the success of the first Macs, was getting ready for the next step and he went to Cray Computer to buy one (probably to help design the PowerPC?).
    Anyway, Cray Computers were not just sitting on the shelf waiting to be sold, so it seems Jobs created an altercation and demanded to see the manager about getting one, so they called Seymour down to the lobby. Steve introduced himself and said words to the effect of “I’d like to use a Cray to design the next Apple Computer”. Seymour replied “Thats great. I used an Apple Computer to design my Cray”.

    • by tlhIngan (30335) <slashdotNO@SPAMworf.net> on Tuesday August 31, 2010 @04:39PM (#33429664)

      Seems Steve Jobs, upon the success of the first Macs, was getting ready for the next step and he went to Cray Computer to buy one (probably to help design the PowerPC?).
      Anyway, Cray Computers were not just sitting on the shelf waiting to be sold, so it seems Jobs created an altercation and demanded to see the manager about getting one, so they called Seymour down to the lobby. Steve introduced himself and said words to the effect of "I'd like to use a Cray to design the next Apple Computer". Seymour replied "Thats great. I used an Apple Computer to design my Cray".

      Not sure about your quote, but Apple did have a Cray. That's why they're address is "1 Infinite Loop" - the joke was the Cray was so fast, it ran an infinite loop in seconds.

      Then again, a quick Google came up with these links
      http://www.clock.org/~fair/computers/sgi-cray.html [clock.org] (it was used for a supercomputing project, and it was Sculley)
      http://www.thocp.net/biographies/cray_seymour.htm [thocp.net] claims the quote is "When told that Steve Jobs bought a CRAY to help design the next Apple, Seymour Cray said, "Funny, I am using an Apple to simulate the CRAY-3." http://www.spikynorman.dsl.pipex.com/CrayWWWStuff/Cfaqp3.html#TOC23 [pipex.com] seems to have a more detailed version of the Apple-Cray connection.

      I guess the next question is - why didn't Microsoft have one?

      • by Caerdwyn (829058)

        This may be just a story or it may be the real deal. The folks I heard it from were in a position to know, so I favor the latter. Perhaps someone out there can be more certain on this.

        Apple's Cray was used, among other things, to simulate plastic-flow and cooling for injection molding tasks. This was necessary because injection-molding complex shapes such that they did not have internal cleave-planes from cooling molten plastic from one injector coming into contact with cooling molten plastic from another

  • I looked at all of the pictures, and there were some missing parts: teletype, card reader, tape drives, and a few guys in white lab coats. Oh, and the glass box surrounding everything so the lower beings can see the magic, but have no ability to actually touch the equipment. I think an aquarium would be good, build an elevated floor with removable 2x2 scale foot panels. You would have to cut two little windows in it, one for the card decks coming in, and one for the card decks wrapped with printouts comi
  • Building a binary-compatible Cray clone with an FPGA board: Awesome!

    Putting it in a case that's...sorta roughly the same shape as a Cray-1: LAME!

    Here's a MUCH better scale Cray-1 case mod housing two x86 PCs:

    http://www.bit-tech.net/modding/case-mod/2010/07/28/cray-1-by-daryl-brach/7 [bit-tech.net]

Do not simplify the design of a program if a way can be found to make it complex and wonderful.

Working...