Slashdot Mirror


Virginia Tech Announces Supercomputer Plans

CousinVinnie writes "Previously noted in this Slashdot story, the administration of Virginia Tech has announced they're puchasing 1100 G5's (another story) in hopes to build a top-10 supercomputer by October 1. Tech will be spending $5.2 million over five years on the project, which should help it pull in more research money." Maybe VT can use the new computers to beef up their web site.

419 comments

  1. This is quite cool but... by WinterSolstice · · Score: 4, Interesting

    Does anyone know who else was considered for this contract? I'd love to see the arguments for the different platforms!

    -WS

    --
    An operating system should be like a light switch... simple, effective, easy to use, and designed for everyone.
    1. Re:This is quite cool but... by EvilTwinSkippy · · Score: 5, Insightful
      Itanium: er, um, we have a new architecture! I think RedHat has a port to it.

      G5: We have a PowerPc system that has been extended to use 64 bits. Your old software will run. Your new software will run faster. We have MacOSx, BSD, and Linux available, natively compiled.

      There is also something to be said for the G5's parallel memory busses. It divides the ram in half, each half feeding 32 bits of the processor. You could theoretically keep your instructions on one side and data on the other, and pipeline the snot out of it.

      --
      "Learning is not compulsory... neither is survival."
      --Dr.W.Edwards Deming
    2. Re:This is quite cool but... by Kalak · · Score: 5, Interesting

      Dell and HP were considered, and Apple won based to a large degree on delivery date. There are more issues to computing than benchmarks, and in the issue of deliverability, Apple won. If you RTFA on the CT, they say it was on the speed and memory of the G5, but the geek grape vine, and hints from the Roanoke times article said availability to get it up in time to make the next top 500 comuter listing we big factors.

      Both Dell and HP have recently announced large clusters, so that may be why they were unable to deliver in time.

      --
      I am, and always will be, an idiot. Karma: Coma (mostly effected by .hack)
    3. Re:This is quite cool but... by BWJones · · Score: 4, Interesting

      Does anyone know who else was considered for this contract? I'd love to see the arguments for the different platforms!

      Well, considering that the G5 has many of the architectural features of those $40k SGI Octanes that I purchased a few years ago, I would consider that pretty impressive. In short, Apple designed the G5 machines with completely independent busses, so that saturating say an I/O bus will not have any effect on the throughput of say memory to CPU. They are pretty impressive and I can see why many folks who are currently using the Octanes etc... would want new G5's.

      So, you have a UNIX box with true plug and play, 64-bit, nice GUI, full CLI access, Firewire, USB, REALLY nice archetecture etc...etc...etc... All that makes for a pretty convincing argument for clusters moving to the G5's

      --
      Visit Jonesblog and say hello.
    4. Re:This is quite cool but... by heh2k · · Score: 2, Informative
      There is also something to be said for the G5's parallel memory busses. It divides the ram in half, each half feeding 32 bits of the processor.

      multibanked ram is nothing new. it's been around since the 486 days for consumers (iirc), and much earlier in big machines, i'm sure. afaik, most mobosthese days are at least 128bits wide. my alpha (up2000+) is 256.

      You could theoretically keep your instructions on one side and data on the other, and pipeline the snot out of it.

      which would just be slower. 8)

    5. Re:This is quite cool but... by hackstraw · · Score: 3, Insightful

      Itanium: er, um, we have a new architecture! I think RedHat has a port to it.

      I admin a cluster of Itaniums. They are very fast and have not had one single hint of a problem with them.

      To me, something that is completely unacceptable about the G5s for scientific use is that the machines do not support ECC memory!

      My users run up to 5 days at a time across 8-10 processors, and its not cool to get a wrong answer after that run, and possibly never know about it.

      I personally would not care to admin a cluster of Macs. I think they are excellent machines, but not for science.

    6. Re:This is quite cool but... by chill · · Score: 0, Insightful

      Itanium: er, um, we have a new architecture! I think RedHat has a port to it.

      G5: We have a PowerPc system that has been extended to use 64 bits. Your old software will run. Your new software will run faster. We have MacOSx, BSD, and Linux available, natively compiled.


      AMD Opteron would be another item. Your old software will run. Your new software will run faster. We have BSD and Linux available, natively compiled.

      Did we mention they're cheaper than the Apple parts? ($1,500 per machine, retail when using online configurators from various suppliers.)

      There is also something to be said for AMD's HyperTransport bus, with it's ability to minimize latency between interconnects and optimizing I/O, like when the CPU is talking to the network card.

      * * *

      While the G5 is a lovely machine, it isn't lightyears ahead of everything else like it has been portrayed. And given Apple's (okay, Motorola's & IBM's) much slower timetable in upgrading CPU speeds, it won't stay "faster PC on the planet" for very long.

      --
      Learning HOW to think is more important than learning WHAT to think.
    7. Re:This is quite cool but... by Anonymous Coward · · Score: 1, Funny

      ECC is dying...

    8. Re:This is quite cool but... by BoomerSooner · · Score: 1

      How fast is the Opteron again?

    9. Re:This is quite cool but... by Krach42 · · Score: 3, Informative

      *ahem* The PowerPC architecture wasn't extended to support 64-bit. It was the IBM POWER architecture that was "extended" to support 32-bit from 64-bit. The original PowerPC designs were designed to be executably compatible with the POWER architecture.

      This is ENTIRELY unlike the x86 architecture, which has been extended to support 32-bit from 16-bit, and now is being extended YET AGAIN to support 64-bit from 32-bit from 16-bit.

      --

      I am unamerican, and proud of it!
    10. Re:This is quite cool but... by BWJones · · Score: 4, Funny

      multibanked ram is nothing new. it's been around since the 486 days for consumers (iirc), and much earlier in big machines, i'm sure. afaik, most mobosthese days are at least 128bits wide.

      Yep, my old Macintosh 9600 had a 128-bit wide memory bus if you used identical ram in each of the 12!! RAM slots.

      --
      Visit Jonesblog and say hello.
    11. Re:This is quite cool but... by Anonymous Coward · · Score: 1, Interesting

      Just curious. Do you log your memory errors, and if so what is the error frequency?

    12. Re:This is quite cool but... by nullard · · Score: 2, Informative

      There is also something to be said for AMD's HyperTransport bus

      Yeah,HyperTransport is pretty cool. Too bad the G5 doesn't have it, right?

      --


      t'nera semordnilap
    13. Re:This is quite cool but... by Frobozz0 · · Score: 3, Insightful

      Apparently the PHD's at Virginia Tech disagree with you 5.2 million ($) times. Or 1,100 times, depending on your view of the world...

      --
      "Politicians find new names for institutions which under old names have become odious to the people."
    14. Re:This is quite cool but... by stratjakt · · Score: 2, Informative

      Parity errors are extremely frequent, the average desktop PC suffers about one a week if it's run all the time. When the difference between 1 and 0 is a handful of electrons, its not surprising.

      The reason its not a big deal in the desktop world, is that you rarely notice those errors. Depends what got changed. Maybe a pixel in a bitmap got a little redder. Chances are it will happen in unused memory when the computer is idle. Modern PCs are idle most of the time, anyways.

      But when you start demanding 100% of the CPU, Bus and Memory in a high usage environment and demand complete accuracy and 5 9's of uptime, it becomes a huge issue.

      --
      I don't need no instructions to know how to rock!!!!
    15. Re:This is quite cool but... by hackstraw · · Score: 2, Informative

      Just curious. Do you log your memory errors, and if so what is the error frequency?

      Actually, yes. And they are mailed to me nightly. I usually get 1 or 2 correctable errors a week.

    16. Re:This is quite cool but... by WatertonMan · · Score: 2, Informative
      A couple points both for the parent and a few comments to the parent.

      Apple uses HyperTransport. It is in the custom chip they designed and IBM manufactures for them. It is only for the memory controller though. The FSB bus on the G5/970 is IBM's Elastic Bus. It is very similar to HyperTransport but not technically the same thing.

      There's an excellent discussion at Ars on this.

    17. Re:This is quite cool but... by be-fan · · Score: 3, Informative

      Read the tech doc and see where that Hypertransport link is. In the Opteron, its between CPUs, which allows for a very low latency pseudo-NUMA setup. On the G5, its between the Northbridge and the Southbridge. Even low end Athlons these days have Hypertransport links there.

      --
      A deep unwavering belief is a sure sign you're missing something...
    18. Re:This is quite cool but... by repetty · · Score: 1

      >> Actually, yes. And they are mailed to me nightly.
      >> I usually get 1 or 2 correctable errors a week.

      And how many undetected errors do you log each week?

      --Richard

    19. Re:This is quite cool but... by tonyhill · · Score: 1

      I agree that ECC is very much needed. How do you know that Apple isn't putting ECC memory into these boxes? Is the motherboard chipset incapable of handling ECC memory?

      Really, I am curious.

      Afterall, it does seem like Apple's doing some customiziation beyond what we can order at the Apple Store (Infiniband interconnect).

      Tony

    20. Re:This is quite cool but... by SewersOfRivendell · · Score: 4, Informative
    21. Re:This is quite cool but... by Anonymous Coward · · Score: 0

      Try getting a vendor (any reputable vendor), to agree to deliver a couple of thousand chips worth of itanium blades within a 6 month window...

    22. Re:This is quite cool but... by Faceprint · · Score: 1

      We've got an Operon cluster too. It's just not as big, only 600+ nodes, if memory serves. It may still be only Athlons right now, I forget if the upgrade has happened yet, but it's definitely in the works.

    23. Re:This is quite cool but... by Kenneth+Stephen · · Score: 0, Troll

      Probably not. More likely, some money-wad holding executive in the administration who loved Macs, disagreed with the Ph.d's 5.2 million times. The probability is that you are a pimply faced teenager who hasnt yet had a chance to learn how the real world works. Hence your statement.

      --

      There is no such thing as luck. Luck is nothing but an absence of bad luck.

    24. Re:This is quite cool but... by hackstraw · · Score: 2, Funny

      And how many undetected errors do you log each week?

      infinity - 7.3 I think.

    25. Re:This is quite cool but... by Textbook+Error · · Score: 1

      How do you know that Apple isn't putting ECC memory into these boxes? Is the motherboard chipset incapable of handling ECC memory?

      The current G5 motherboards do not support EEC memory. I asked about support for EEC at WWDC, and the reply was that they were definitely thinking about that and similar high-reliability features.

      I'm quite sure that things like EEC and G5 Xserves are on the drawing boards, but realistically they can't do everything at once - and for Apple, the first rev of these machines is a replacement for their existing desktop models (big sales like this are impressive for PR, but don't sustain a company the size of Apple).

      --

      Nae bother
    26. Re:This is quite cool but... by Textbook+Error · · Score: 1

      Or ECC even, too much time in Europe...

      --

      Nae bother
    27. Re:This is quite cool but... by BigBadBri · · Score: 1
      Who ever runs a simulation just once?

      If the computer is fast enough, why not run each sim twice, and rerun if the results disagree?

      ECC is good for reliablility, and I certainly wouldn't run a nuclear power plant on a non-ECC machine, but if you've got 1100 processors to play with, then just run everything twice.

      If you think back 20 years, people used to analyse their problems carefully to minimise computer time, and all the runs were probably justified.

      I doubt whether half the runs these days are strictly necessary - better experimental design could probably eliminate a hell of a lot of computer time.

      --
      oh brave new world, that has such people in it!
    28. Re:This is quite cool but... by penguin7of9 · · Score: 0, Troll

      Apparently the PHD's at Virginia Tech disagree with you 5.2 million ($) times. Or 1,100 times, depending on your view of the world...

      Most likely, they got a special deal from Apple, or maybe some other deal was tied in with this and they get a discount on the other deal.

      There are thousands of x86/Linux-based Beowulf clusters. There are hundreds of NT-based compute clusters. So, a few universities who also use Macs for compute clusters--not exactly surprising. Maybe they have a good reason why it makes sense in their environment or maybe they are simply making a mistake--it happens.

      I see little to recommend a Mac cluster over a Linux cluster. The G5 processors are pretty nice, but in terms of bang-for-the-buck, they aren't anything special. And while OS X makes a pretty desktop operating system, it is very rarely used for compute clusters so there isn't a lot of software for it for that purpose.

    29. Re:This is quite cool but... by Glock27 · · Score: 1
      The G5 processors are pretty nice, but in terms of bang-for-the-buck, they aren't anything special.

      Perhaps not compared with blade servers (like the upcoming G5 XServes).

      However, compared with Opteron workstations, the G5s are a great deal. :-)

      --
      Galileo: "The Earth revolves around the Sun!"
      Score: -1 100% Flamebait
    30. Re:This is quite cool but... by Greedo · · Score: 4, Funny

      I think they are excellent machines, but not for science.

      Well, like ...

      One night, I was processing these gene sequences on my Itanium cluster, you know? When all of a sudden it went berserk, the screens started flashing, and, like, the whole result set just disappeared. All of it. And it was a good result set! I had to cram and resequence it really quickly. Needless to say, my rushed thesis wasn't nearly as good, and I blame those Itaniums for the funding didn't get, and tenure I lost.

      I'm happy to report that my sister and I now share an Apple G5 cluster. It's a lot nicer to work on than my old Itanium farm was, it hasn't let me down once, and my papers have all been really good.

      Thanks, Apple.

      Ellen Feiss, PhD

      --
      Tuus crepidae innexilis sunt.
    31. Re:This is quite cool but... by stilwebm · · Score: 1

      There are more issues to computing than benchmarks, and in the issue of deliverability, Apple won. ...

      hints from the Roanoke times article said availability to get it up in time to make the next top 500 comuter listing we big factors


      This is significant because both Virgina Tech and Apple gain from an early delivery. Apple gets to promote it's G5s as being part of a top 10 super computer, and Virgin Tech gets to promote itself both to prospective students and for the purpose of research financing. From the stories of dual G5 users having their ship date delayed, I believe Apple bent over backwards to get a favorable delivery date.

    32. Re:This is quite cool but... by zerocool^ · · Score: 4, Informative

      identical ram in each of the 12!! RAM slots

      Dude, look at an old sparc sometime. Sparc 1/1+/2 had 16 ram slots, circa 1990. Of course, you had to fill 4 at a time. The max is 128 MB i think.

      ~Will

      --
      sig?
    33. Re:This is quite cool but... by ProfessionalCookie · · Score: 2, Informative

      I think they are excellent machines, but not for science.

      Of course WinTel is the ultimate science platform. It's not like intel has a history of CPUs that can't do math. It's not like windows doesn't have a history of wild instability.

      I think Apple's are great machines too- but I think you're a troll.

    34. Re:This is quite cool but... by Anonymous Coward · · Score: 0

      Since they've barely sold a couple thousand Itaniums in total, that request would be taken very seriously.

    35. Re:This is quite cool but... by Valar · · Score: 1

      Would that be Pointy Haired Doctors? Seriously though, when will people get it. The g5 is a pretty damned good processor, and the architecture around it is definitely better than that of the p4. Full speed mem bus, how novel!

    36. Re:This is quite cool but... by Anonymous Coward · · Score: 0

      Who ever runs a simulation just once?

      Eeerh... the point is that you want to run them multiple times with different random seeds. A computer should be deterministic (think Turing).

      If the computer is fast enough, why not run each sim twice, and rerun if the results disagree?

      Supercomputers are never "fast enough". Are you really suggesting that scientists should run all simulations twice just because Apple doesn't use ECC? In all fairness, if that's the case they should only report half the LINPACK performance to top500.

    37. Re:This is quite cool but... by AllMightyPaul · · Score: 1

      The infiniband is being added by Tech. That doesn't come with it.

    38. Re:This is quite cool but... by Anonymous Coward · · Score: 0

      Where did you get 1100 G5/ Dual 2ghz when VT just got their's? Come on. If you are going to exaggerate, make it possible.

    39. Re:This is quite cool but... by Gizzmonic · · Score: 1

      I thought this was very funny.

      But I think it would be funnier if you made it a cluster of Pentium Pro's running Xenix.

      --
      (-1, Raw and Uncut is the only way to read)
    40. Re:This is quite cool but... by Anonymous Coward · · Score: 0

      Does it matter?

      This is "JAC", Just Another Cluster.

      Putting COTS in a rack is about as exciting as
      watching paint dry. G5 or Intel, it doesn't
      really matter and will be obselete the moment
      it is installed.

    41. Re:This is quite cool but... by bspath1 · · Score: 3, Informative

      Actually, PowerPC was designed as a 64 bit architecture from the ground up. The first implementations were 32 bit, although the PPC 620 was 64 bit AFAIR. The 601 was a hybrid PowerPC/POWER chip which supported both ISAs. The 603 and 604 chips were 32 bit PowerPC implementations and lacked the deprecated POWER instructions which were supported by the 601. IBM also went on to extend POWER to POWER2, POWER3 and a 64 bit implementation of POWER for their mainframes (RS64? and AS/400?). POWER4 is a unification of the 32 bit POWER and 64 bit POWER architectures which is also fully PowerPC 32 and 64 bit compatible. Note: the 970 is based on POWER4.

    42. Re:This is quite cool but... by jandrese · · Score: 1

      I don't know. That seems like a lot of parity errors to me. I used to run parity memory on my old PII-400 and I saw (I think) 3 SBEs in 5 years. On my new machine I decided to go nonparity because it just didn't seem to be worth the cost. Also, the Nforce chipsets apparently don't support parity memory according to the manual.

      --

      I read the internet for the articles.
    43. Re:This is quite cool but... by _iinc · · Score: 1

      my quadra 950 also has 16 ram slots, all of the old 30-pin variety. I don'tknow what the max is but it's sitting at 100MB right now.

    44. Re:This is quite cool but... by Anonymous Coward · · Score: 0

      Remember that compared to your old PC, newer memory technologies use larger amounts of smaller memory elements. Smaller memory elements make it more likely that a single cosmic ray or disturbance in the force or whatever will change a bit. So advances in memory technology make memory more and more unreliable. At some point ECC will be a necessity... and for commodity systems that point would probably be when memory errors demonstrably overwhelm Windows software errors in frequency. We have a decade or so to wait for that...

    45. Re:This is quite cool but... by Anonymous Coward · · Score: 0

      you looked at http://www.top500.org lately? ;)

    46. Re:This is quite cool but... by Hoser+McMoose · · Score: 1

      Why not save yourself 2.7 million dollars and get the same computing power while only running the tests ONCE!

      If they're spending that much money to buy that much equipment, you better believe that they expect to USE it! Running tests twice might be fine for you or me on our home computers which are going to be sitting idol 99% of the time anyway, but duplicating everything you do every time when your building a supercomputer is just wasting money, and a lot of it.

      Do you run a simulation once? Of course not! But what's the point in running the simulation twice with the exact same input data and/or random seed? That doesn't help prove anything except that it works/doesn't work for one very specific case.

    47. Re:This is quite cool but... by Hoser+McMoose · · Score: 2, Informative

      No, the G5 does NOT have Hypertransport, at least assumign that you are using Apple nomenclature of calling the IBM PowerPC 970 CPU the "G5" and the machine itself a "PowerMac".

      The PowerMac G5 machines use Hypertransport to connect their motherboard chipsets together. This is nothing new, the XBox does this as well. Hypertransport is a very good, low-cost, high performance solution for connection chips together directly. However, connecting motherboard chipsets together doesn't do much of anything for getting data to/from the processor itself.

      IMO the IBM PowerPC 970 (aka the G5) is actually not a bad chip to power a supercomputer. It seems to have quite a lot of processing power for a reasonable price and decent thermal characteristics (when you have a LOT of processors in small space, low power consumption is a very good thing, this is where most of Cray's real innovation was). However, the PowerMac's seem to be rather weak for this sort of application because of Apple's chipset, which is really a desktop/workstation chipset. The I/O bandwidth isn't all that impressive (1.6GB/s in each direction is the absolute best that can be managed) and latency should be even worse. Getting to a PCI-X card involves going through two separate chips and three buses. Compared to some of the proposed Opteron solutions which will hang networking hardware right off the processors Hypertransport buses (no chips in the middle and only a single bus with 3.2GB/s of bandwidth in each direction), the chip I/O looks pretty weak.

      My guess is that this deal is mainly due to the fact that Apple gave them a real good deal. Given the rather high cost of the networking equipment that will be needed, VT is probably getting their 1100 servers for next to nothing.

    48. Re:This is quite cool but... by Krach42 · · Score: 1

      Well put, but in any case, it's definately not extended to support 64-bit, but supported it from the start.

      --

      I am unamerican, and proud of it!
    49. Re:This is quite cool but... by saltyboy · · Score: 0

      ahum... I was poking fun at the freelance gig troll... - duh.

    50. Re:This is quite cool but... by Khlatu_Barada_Nicto · · Score: 0
      sitting idol 99% of the time

      ...in the midnight hour, she cried "More! More! More!".....

    51. Re:This is quite cool but... by Anonymous Coward · · Score: 0

      Hey, this AC thought it was funny. Too bad you shot your karma all to hell...:-(

    52. Re:This is quite cool but... by class_A · · Score: 0

      The XServe is not a blade server. It's a normal 1U server.

    53. Re:This is quite cool but... by Snaller · · Score: 1

      I've been sitting here at my freelance gig in front of a Mac Cluster (1100 G5/Dual 2ghz with(4*1100) Gigs of RAM) for about 20 minutes now while it attempts to copy a 17 Meg file from one folder on the hard drive to another folder. 20 minutes. At home, on my Pentium Pro 200 running NT 4, which by all standards should be a lot slower than this Mac cluster, the same operation would take about 2 minutes.

      I doubt it. What matters when you copy is if its from one harddrive to another, that's fast. If you copy from and to the same disk, that's slow because the head has to move back and forth on the same disk.

      --
      If Google really cared they would fix Android Chrome to reflow text, instead of discriminating
    54. Re:This is quite cool but... by Anonymous Coward · · Score: 0

      YHBT. HAND.

      (This is a very old one, updated periodically by various trolls.)

  2. Yikes.. by knghtrider · · Score: 2, Interesting

    1100 G5's...that should corner the market for about a week...and give Apple a small boost to it's bottom line..

    --
    In America today you can murder land for private profit. You can leave the corpse for all to see, and nobody calls the c
  3. cray by pheared · · Score: 3, Funny

    Burns: [throws his glass at Homer]
    You call this Postum?
    [bashes a 5-feet high pile of paper]
    Burns: You call this a tax return?
    [bangs a CRAY with his cane]
    Burns: You call this a supercomputer?

  4. Performance comparisons... by EvilTwinSkippy · · Score: 5, Funny

    The comparison is like Apples to Oranges. Most people end up asking "Orange you going to build a beowulf cluster of those Apples?"

    --
    "Learning is not compulsory... neither is survival."
    --Dr.W.Edwards Deming
    1. Re:Performance comparisons... by hackstraw · · Score: 3, Funny

      I hear this comparison of apples and oranges all the time.

      Aren't both fruits that grow from trees?

      How about comparing sea water and comets?

    2. Re:Performance comparisons... by tmasssey · · Score: 2, Insightful
      Arent't they both made up of water?

    3. Re:Performance comparisons... by BetterThanCaesar · · Score: 1

      The point is of course that they are both fruit, wherefore one would think that they are comparable. Yet, they are very different in nature and cannot be compared.

      --
      "Stop failing the Turing test!" -- Dilbert
    4. Re:Performance comparisons... by jeremy_a · · Score: 1

      I keep this link in my bookmarks for times like this: Apples and Oranges -- A comparison

    5. Re:Performance comparisons... by SeanAhern · · Score: 1

      My favorite involves a comparison between "night" and "1."

    6. Re:Performance comparisons... by Anonymous Coward · · Score: 0

      When comparing what? If comparing market prices per kg it's pretty simple and fair.

  5. Maybe... by Ieshan · · Score: 4, Funny

    Maybe Apple will use this G5 cluster against a single-processor itanium to show that, yes, they ARE the fastest personal computer!

    The only problem will be finding a desk big enough to fit the guys...

    1. Re:Maybe... by jmenezes · · Score: 4, Funny

      Well, reguardless of the performance of the itanium vs. the G5 cluster, at least you'll be saving on your electricity bill by going with the 1100 G5s....

      --
      Stop over-analyzing your analizations
    2. Re:Maybe... by EvilTwinSkippy · · Score: 2, Funny
      This is college. I'm sure some engineering students would have adapted the heat sinks on the Itanium into a whisky still or an expresso machine.

      (Sigh) Quitters.

      --
      "Learning is not compulsory... neither is survival."
      --Dr.W.Edwards Deming
    3. Re:Maybe... by Phroggy · · Score: 1

      The only problem will be finding a desk big enough to fit the guys...

      It'll also be one of the five largest desks in the world.

      --
      $x='S24;r)>63/* h@<5+oZ)32"5cz';$me='phroggy'x$];
      $x=~y+ -xz+\0-Tx+;print$_^chop$me for split'',$x;
    4. Re:Maybe... by penguin7of9 · · Score: 1

      I don't know about Itanium, but the power dissipation of the G5 and the AMD Opteron are not all that different.

      PPC used to have a big advantage over the x86 architecture in the past, both in terms of performance and power, but those days are behind us: PPC and x86 designers are designing in the same space and using similar techniques and skills; the actual instruction set is just a minor detail.

    5. Re:Maybe... by Lars+T. · · Score: 1

      Yeah, a factor of 2 (okay, 1.9) is "not all that different".

      --

      Lars T.

      To the guy who modded me down from perfect to terrible Karma - Apple haters still suck

    6. Re:Maybe... by AllMightyPaul · · Score: 1

      Not to worry. VT has its own power station on campus.

  6. Re:repeat story by Ratphace · · Score: 1


    Yeah, and his story says it was reported earlier, if you read it...

  7. Re:Virginia Tech by Anonymous Coward · · Score: 0

    They need the power to play duck hunt, bitch. Shut up.

  8. Overpriced G5s by Brahmastra · · Score: 0, Insightful

    Overpriced G5s for what? Instead of buying 1 G5, you could set up 4 top-of-the-line AMD processors, cluster it using Myrinet.. and all that would probably cost about as much or slightly more than G5. Apple may have a good processor that is one of the highest performers... but it costs way too much to be worth it.

    1. Re:Overpriced G5s by SeanAhern · · Score: 4, Insightful

      Wait a minute - you're complaining about the cost of a G5, but go on to suggest they buy a Myrinet, a rather expensive interconnect. Something doesn't compute here.

    2. Re:Overpriced G5s by Brahmastra · · Score: 2, Insightful

      Was just trying to point out that even with an expensive interconnect such as Myrinet, the economics of apple just doesn't work out. But then, even if you are using a cluster of G5s, to get any reasonable super-computing power out of it, you would need a low latency (expensive) interconnect.

    3. Re:Overpriced G5s by Anonymous Coward · · Score: 1, Interesting

      Well, there is the small matter of thermal dissipation--heat to polysyllabically-challenged types like yourself. Those AMDs you're babbling about would cook themselves; the traditional advantage of PowerPC processors for clustering is in the fact that they run MUCH cooler.

      Even if you could make the AMD-based cluster run, you'd have spent so much on cooling that the G5s would end up being cheaper. Just ask DOD...

    4. Re:Overpriced G5s by SeanAhern · · Score: 4, Insightful

      to get any reasonable super-computing power out of it, you would need a low latency (expensive) interconnect.

      Well, that very much depends on what type of computing you're doing. Some scientific computing is more tolerant of high-latency environments and would rather have the bandwidth.

      I can't seem to find the quote from any of the articles right now, but VT is planning on using an Infiniband interconnect from Mellanox. While I don't know the relative price points, they are touting the fact that this is a high-speed interconnect that's faster than Myrinet or Quadrics at a fraction of the cost. I can't say for sure, since the Infiniband cluster we're helping to build at Stanford is not yet assembled.

      This should be interesting to watch. I'll be very interested to see the $/gigaflop ratio for VT's cluster (though that doesn't have a bearing on the interconnect).

    5. Re:Overpriced G5s by 11223 · · Score: 5, Insightful
      OK, a few relevant points here:

      • Myrinet sucks. No, really. It does. It eats CPU when sending data. It generally does not perform as well as Infiniband. It has higher latency. Once you start using this for real work (outside of what are known as "embarassingly parallel" problems, which are fine with 10baseT), those factors will play into your performance to a huge degree.
      • AMD fanboys, take note: The G5 does have superior floating-point hardware, for either double precision superscalar or vector single precision work. If they're doing floating point the G5 is a clear win.
      • The memory bus on the G5 is a bit better than on the Opterons - especially once you start doing threaded work, the dual unidirectional buses essentially allow cache transfers at the same time as memory transfers, and a whole bunch of other possibilities.
      • Lastly, what are you smoking? The only way an AMD becomes competitive with a G5 (machine to machine here) is if you build the AMD yourself and leave out the stuff the G5 has in it. Are you suggesting that Virginia Tech build 1100 Opteron, no, 4400 Opteron systems (you said 4x, not me!) themselves? That's crazy. They want somebody else to build and test the machines, and that somebody to be responsible when they fail. Of course such a real-world advantage has little to do with the bubble most /.ers live in, because they build their Athlons themselves. Perhaps the reality distortion field merely applies to introducing reality to people who have never seen it before.
    6. Re:Overpriced G5s by switcha · · Score: 2, Insightful

      How about resale? When the projects wind down and things need upgrading, they can get maybe over half a mil' in return for offloaded desktops (or at least scrap aluminum), as opposed to 57 cents for a bunch of beige schrapnel.

      --
      You know what? ... A little club soda *did* get that out!
    7. Re:Overpriced G5s by Brahmastra · · Score: 3, Insightful

      Myrinet eats CPU when sending data?? You must be using the suckiest driver and firmware possible. And Infiniband has lower latency than Myrinet???? Infiniband is a combination of protocol and hardware and actually Infiniband has slightly higher latencies than the best MPI implementations on Myrinet. Myrinet is a just a piece of hardware. You can write firmware in Myrinet to do almost everything in the Lanai processor present in the card itself, without consuming any CPU cycles. The performance you get out of Myrinet entirely depends on the libraries you are using.

    8. Re:Overpriced G5s by Anonymous Coward · · Score: 0

      >>The only way an AMD becomes competitive with a G5 (machine to machine here) is if you build the AMD yourself and leave out the stuff the G5 has in it.

      Yes, because it's very important to have firewire and itunes on your supercomputer.

      >>They want somebody else to build and test the machines, and that somebody to be responsible when they fail.

      And there are many companies already building 2 and 4 processor Operton boxes right now. What's your point?

      >>Perhaps the reality distortion field merely applies to introducing reality to people who have never seen it before.

      I can clearly see who is really affected by the RDF(TM).

    9. Re:Overpriced G5s by Tiosman · · Score: 1

      How does Myrinet "eats CPU when sending data" ?!? It uses DMA to read/write data from/to host memory. The only thing the CPU is doing is writing 48 Bytes by PIO to post the send, whatever is the size of the message to send.

      So either this is a flame bait or you really have no idea what you are talking about. I think I do as I write code for Myrinet firmware.

    10. Re:Overpriced G5s by Smurf · · Score: 1
      >>They want somebody else to build and test the machines, and that somebody to be responsible when they fail.

      And there are many companies already building 2 and 4 processor Operton boxes right now. What's your point?

      The grandfather of your post said that you could buy 4 top-of-the-line AMD processors for the cost of a G5. (Of course, he meant complete systems, not only the processors).

      There are several companies building 2 and 4 processor Opterons, but NOT for costs such that each G5 can be replaced by 4 Opterons.

      That's the point of the parent post: you can get extremely cheap AMD boxes (specially if you build them yourself). And you can get multiprocessor AMD machines fully supported and guaranteed by serious companies. But you can't get both on the same package.

      The cheap ones are not an option for a serious massive project like Virginia Tech's. The classy ones are (price ways) on the same ball park as the G5 (admittedly cheaper but by no means by a factor of four).

    11. Re:Overpriced G5s by ruiner5000 · · Score: 1
      I want to see proof of any of this.

      AMD fanboys, take note: The G5 does have superior floating-point hardware, for either double precision superscalar or vector single precision work. If they're doing floating point the G5 is a clear win.
      The memory bus on the G5 is a bit better than on the Opterons - especially once you start doing threaded work, the dual unidirectional buses essentially allow cache transfers at the same time as memory transfers, and a whole bunch of other possibilities.
      Lastly, what are you smoking? The only way an AMD becomes competitive with a G5 (machine to machine here) is if you build the AMD yourself and leave out the stuff the G5 has in it. Are you suggesting that Virginia Tech build 1100 Opteron, no, 4400 Opteron systems (you said 4x, not me!) themselves? That's crazy. They want somebody else to build and test the machines, and that somebody to be responsible when they fail. Of course such a real-world advantage has little to do with the bubble most /.ers live in, because they build their Athlons themselves. Perhaps the reality distortion field merely applies to introducing reality to people who have never seen it before.

      Apple won't give out review systems to real computer hardware sites, and so we have no proof I'm aware of. If you do I want to see your head to head tests of G5 and Opteron in single, dual, quad, and clusters since you obviously have proof and have tested all of this. You can now get two flavors of Windows, 3 Linux flavors, and FreeBSD is around the corner for AMD64. Where are all the 64 bit G5 operating systems again? Apache is ported, mysql is ported, Seti@Home, etc, etc. Where are all the OSX 64 bit program ports?

      --
      ignorance is bliss. googlefiberatx.com
    12. Re:Overpriced G5s by RyuuzakiTetsuya · · Score: 1

      Yeah, but what kind of framerate will this giant G5 cluster see? My Athlon at home...

      (I'm kidding, I don't use an Athlon at home, I use a P3 Tualatin.)

      --
      Non impediti ratione cogitationus.
    13. Re:Overpriced G5s by trollabyte · · Score: 1
      Myrinet sucks... ...It eats CPU...
      1. That may have more to do with mpich than myrinet. In any event, my applications run much faster with myrinet (upwards of 2x), I don't care that the cpus are pegged. Running with IB may be great, but there isn't exactly a large set of commercial applications out there that support it. Not everybody runs in a research environment where you're writing your own code, or doing your own linking. You try getting a vendor to release libraries...
    14. Re:Overpriced G5s by UnixRevolution · · Score: 1

      I did build my athlon...

      but i *bought* my ibook. Warranty, yo. sometimes it's worth it. plus OS X is uber-sweet.

      --
      You like your new Mac more than you like me, don't you, Dave? Dave? I asked...She said Yes.
    15. Re:Overpriced G5s by Hoser+McMoose · · Score: 1

      That "traditional advantage" doesn't apply here because the power consumption of IBM's PowerPC 970 and the AMD Opteron are nearly identical.

      Exact numbers are few and far between on both sides (neither company has a publicly available document that describes their exact power consumption), but most estimates put them as being real close. Both chips consume about 45-50W typical power consumption, and somewhere in the order of 50-60W maximum power consumption. One or the other might be a few watts cooler, but the difference is quite small.

      Both of these chips consume a lot more power than the G4 and both consume a lot less power than the AthlonXP or the Pentium 4 (or the Itanium for that matter, which is the current champion of high power consumption).

    16. Re:Overpriced G5s by Hoser+McMoose · · Score: 1

      Others who know more than I do about Myrinet have already pointed out that you don't seem to know what you're talking about, so I'll skip that.

      Next point, FP performance. Let's see, Apple managed a score of 840 on Spec CFP2000 base on their 2.0GHz PowerPC using GCC. AMD managed a score of 1209 on their 2.0GHz Opteron using Intel's compiler, while IBM managed a slightly lower score of 1172 using GCC. If the G5 is so superior, it sure isn't doing a very good job of showing it.

      The memory bus on the G5 is better? What are you smoking here?! Each Opteron chip has a low-latency integrated memory controller capable of managing 5.4GB/s of bandwidth. Two G5 shares a higher latency, off-chip memory controller capable of 6.4GB/s total. Their bus may allow memory and cache transfers at the same time, the Opteron has a completely separate buses for these tasks! The Opteron also has twice the L1 data cache and twice the L2 cache that the PowerPC 970 has.

      I will, however, agree with you that the Opteron is no cheaper than the G5 if you can get Apple to give you a useful configuration of the boxes for a supercomputer (who the hell needs a high-end gaming video card on each super-computer node? Let alone iTunes!). As for building the machines themselves though, why would they do that? Why not just buy a machine from one of the dozens of companies that are selling dual-processor 1U Opteron servers now. Many can and have built clusters of these already. Speaking of '1U' and all, brings me to one fo the dumbest things about this supercomputer, it's ALL being built usind the damn PowerMac tower cases! Now THAT is just a DUMB idea! 1100 giant desktop cases using bog-standard power supplies and the works, now that is just ridiculous!

    17. Re:Overpriced G5s by kargis · · Score: 1

      Would be madness to build a supercomputer out of Opterons -- and you'd need 4400 of them for it to . . . to do what?

      Specint base Opteron 246 = 1248
      Specint base G5 2 ghz = 800

      Specfp base Opteron 246 = 1209
      Specfp base G5 2 ghz = 840

      These numbers are from spec.org for the opteron and of course apple.com. Now I'm not sure why I need 4400 opterons all of a sudden to equal 2200 of these initially slower machines. And of course, the idea that no one's doing this:

      http://www-1.ibm.com/servers/news/pr_730.html

      ARMONK, N.Y. and TOKYO July 30, 2003 -- IBM today announced that Japan's largest national research organization has ordered an IBM TM Linux(R) supercomputer that when completed will deliver more than 11 trillion calculations per second, making it the world's most powerful Linux-based supercomputer. It is expected to be more powerful than the current Linux cluster currently ranked as the third most powerful supercomputer in the world, according to the widely popular TOP500 List of Supercomputers [1].

      The supercomputer is planned to be integrated with other non-Linux systems to form a massive, distributed computing Grid -- enabling collaboration between corporations, academia and government -- to support various research including grid technologies, life sciences bioinformatics and nanotechnology.

      The system with a total of 2,636 processors will include 1,058 IBM 325 systems, which were introduced today with 2,116 AMD OpteronTM processors. [2]

      At least look into stuff before deciding that it's not competitive -- the opteron is certainly competitive. Sure, it's not a Mac, but it's pretty damned fast, and a quick look online shows a price of ~$3000 -- same as list on the G5.

      Finally, no one knows how well these things will multiprocess -- they're not using linux, they're using OS X -- which is BSD, granted, and should do well, but god only knows what they've done. I actually do think it will be a nice, fast machine, just not necessarily significantly better for specious reasons than a similar opteron machine when the initial base spec marks are so much in favor of the Opteron (800 vs 1200 is 25% last time I checked), and the Opteron has the advantage of having IBM working on making Linux nice, fast, and well-behaved in a cluster setting.

      Kargis

    18. Re:Overpriced G5s by bspath1 · · Score: 1

      The benchmark that this machine is going to be rated against is LINPACK, which involve large matrix calculations and is heavily floating point dependent. An ideal computer would be able to achieve its peak theoretical FP performance in the extreme of an infintely sized matrix calculation under LINPACK, R_peak. This is different from SPEC FP which is a suite of different floating point intensive benchmarks which test various aspects of floating point performance for the CPU, compiler, and overall system.

      AFAIK, the Opteron has 3 asymmetric FPUs where each FPU performs some subset of functionality, and more than one FPU component may need to act in concert to perform other FPU instructions. More to the point, AFAIK, the Opteron is capable of doing 1 floating point multiply and 1 floating point add per cycle.

      The PPC 970 has 2 fully symmetric FPUs each of which can perform any FPU instruction available to the architecture. One of the key features of the POWER ISA family and its descendants is the single cycle FMA instruction which allows each FPU to perform 1 floating point multiply and 1 floating point add per cycle. As a result, for what LINPACK measures, the PPC 970 is capable of twice the theoretical throughput per clock cycle when compared to an Opteron. Since Opterons and PPC 970s run at comparable clock speeds, it's no surprise that it takes twice as many Opterons to equal the performace of a cluster of 970s.

      SPEC FP is irrelevant in this case.

      -Bruce.

  9. Apple ... supercomputer...? by TWX · · Score: 2, Insightful

    Not to sound like a troll, but isn't the Apple a bad machine to use for this? It's big, the fan configuration will make it extrordinarily loud, and it's built to cater to the end user, not to the embedded machine market. Yes, OSX/Darwin does work fairly well, but I'd think that the entire purpose of this computer originally would make it ill-suited to this task.

    Many companies build physically smaller machines that still pack a lot of power, or sell parts to allow someone to design their own layout in a chassis. Remember, individual cases, power supplies, and the like become way overkill in such a large computer, and it would probably be cheaper to convert electricity once for a large section of the computer, supplying 12v, 5v, and 3.3v without each computer converting itself.

    This just seems like the wrong way to do something thats hallmark has been in being cheap.

    --
    Do not look into laser with remaining eye.
    1. Re:Apple ... supercomputer...? by tesmako · · Score: 3, Funny

      Oh nos!1!! The supercomputer will be all loud and stuffs! Whatever shall we do? It will be a pain to play quake on it and ecverything. A monumental failure this.

    2. Re:Apple ... supercomputer...? by TWX · · Score: 1

      "Oh nos!1!! The supercomputer will be all loud and stuffs! Whatever shall we do? It will be a pain to play quake on it and ecverything. A monumental failure this." I nominate tesmako as the first computer operator to have to sit in the 100+ decibel room for the first eight hour shift...

      --
      Do not look into laser with remaining eye.
    3. Re:Apple ... supercomputer...? by gunleiksrud · · Score: 1

      TWX wrote:
      > the fan configuration will make it
      > extrordinarily loud,

      Have you been in the same room as a G5??
      They are suprisngly quiet.

      --
      Gaute Gunleiksrud
      If the Apocalypse comes, beep me!!
    4. Re:Apple ... supercomputer...? by Phroggy · · Score: 5, Insightful

      the fan configuration will make it extrordinarily loud

      Apple specifically engineered these systems to be quiet - the compartments are set up the way they are so they can get maximum airflow with minimal blowing. Just because you think "loud" when you hear nine fans doesn't mean they're actually any louder than anything else. You're spreading FUD.

      it's built to cater to the end user, not to the embedded machine market.

      Perhaps your definition of "embedded" is different from mine, but I'm pretty sure it doesn't fit this application. This is a supercomputer cluster. However you are correct that these machines were designed to be desktop computers. Apparently that's not all they're good for.

      Many companies build physically smaller machines that still pack a lot of power,

      Yeah, so does Apple, but these are faster.

      or sell parts to allow someone to design their own layout in a chassis.

      If Virginia Tech wants to order 1,100 of them, don't you think Apple would be flexible if this was a concern?

      Remember, individual cases, power supplies, and the like become way overkill in such a large computer, and it would probably be cheaper to convert electricity once for a large section of the computer, supplying 12v, 5v, and 3.3v without each computer converting itself.

      This is an interesting point I hadn't considered. Feeding 110v into each of 1,100 individual power supplies can't be as energy or heat efficient as what you describe. However, it's possible that they will actually be doing this - I don't think I've seen it mentioned anywhere.

      Another consideration - apparently VT was pressed for time and they needed something that would be available quickly; Apple was able to deliver quickly. This may explain why they'd be more inclined to use stock off-the-shelf boxes instead of something more customized.

      --
      $x='S24;r)>63/* h@<5+oZ)32"5cz';$me='phroggy'x$];
      $x=~y+ -xz+\0-Tx+;print$_^chop$me for split'',$x;
    5. Re:Apple ... supercomputer...? by hackstraw · · Score: 1

      fan configuration will make it extrordinarily loud

      You must not have ever been in a machine room. They are all pretty noisy and cold.

      OSX/Darwin does work fairly well, but I'd think that the entire purpose of this computer originally would make it ill-suited to this task

      Compute nodes are pretty stripped down. All they really need to do is run one app per cpu. The standard *NIX tools like rsh, rcp, nfs (optional), etc are all you need.

      Remember, individual cases, power supplies, and the like become way overkill in such a large computer, and it would probably be cheaper to convert electricity once for a large section of the computer, supplying 12v, 5v, and 3.3v without each computer converting itself.

      Maybe. I've never heard of such a thing. The whole point of buying a bunch of commodity machines and connect them together is that its cheaper and easier than buying a "real" supercomputer with one or a few boxes.

      This just seems like the wrong way to do something thats hallmark has been in being cheap

      I believe this as well, but for different reasons.

    6. Re:Apple ... supercomputer...? by Karl+Cocknozzle · · Score: 1
      It's big, the fan configuration will make it extrordinarily loud, and it's built to cater to the end user, not to the embedded machine market.

      Have you ever seen/touched/heard a real live G5? The fans go on and off as-needed--all the fans don't run full bore at all times. Sure, there'll be some noise, but ultimately, not an unbearable amount. (Unlike our AMD core server units we've purchased... Those sound like propeller aircraft taking off.)
      --
      Who did what now?
    7. Re:Apple ... supercomputer...? by alfredo · · Score: 3, Interesting

      Apple has won other contracts by being able to act quickly. When the Postal Service needed to set up an intranet, everyone said one to two years. Apple using WebObjects was able to do it in 6 months. At the time the Postal Intranet was the largest in the world.

      --
      photosMy Photostream
    8. Re:Apple ... supercomputer...? by jcr · · Score: 1

      To be precise, the fan speed is variable. They don't just switch on and off.

      -jcr

      --
      The only title of honor that a tyrant can grant is "Enemy of the State."
    9. Re:Apple ... supercomputer...? by Anonymous Coward · · Score: 0
      t's big, the fan configuration will make it extrordinarily loud

      Who cares? Its not as if you are going to be in the same room as these machines

      and it's built to cater to the end user, not to the embedded machine market.


      This i agree with... But still I think the G5 with IBM XL compilers should perform very well in this situation... It also seems to scale nicely...
      Remember, individual cases, power supplies, and the like become way overkill in such a large computer

      Vacancy; Janitor needed
    10. Re:Apple ... supercomputer...? by jceaser · · Score: 1

      Remember, individual cases, power supplies, and the like become way overkill in such a large computer, and it would probably be cheaper to convert electricity once for a large section of the computer, supplying 12v, 5v, and 3.3v without each computer converting itself. This is an interesting point I hadn't considered. Feeding 110v into each of 1,100 individual power supplies can't be as energy or heat efficient as what you describe. However, it's possible that they will actually be doing this - I don't think I've seen it mentioned anywhere.

      I have. Distributed power supplies means than one failure does not bring down the entire computer. I really don't think that at any given time all 100% of these computers will be up and running. A percentage will be down for one reason or another (bad fiber, power supply, harddrive, upgrading). Given the mean fail rate of all hardware, you need 1100 just to make sure 1095 (?) are up and running.

    11. Re:Apple ... supercomputer...? by poot_rootbeer · · Score: 1

      Just because you think "loud" when you hear nine fans doesn't mean they're actually any louder than anything else. You're spreading FUD.

      Furthermore, what does it matter if a supercomputer (I use that term loosely, please don't bother getting pedantic about it) is loud or not? The typical server room IS loud -- if the computer systems aren't making noise, the HVAC systems are. It's a red herring.

    12. Re:Apple ... supercomputer...? by frank_adrian314159 · · Score: 1
      but the execs in charge are pulling in tidy salaries for one reason: to make money for shareholders.

      On the plus side, the aeronautics department can use it simultaneously for a wind tunnel!

      --
      That is all.
    13. Re:Apple ... supercomputer...? by TWX · · Score: 1

      "Maybe. I've never heard of such a thing. The whole point of buying a bunch of commodity machines and connect them together is that its cheaper and easier than buying a "real" supercomputer with one or a few boxes."

      There's a reason why the phone company runs EVERYTING off of 48V. Converting 48V DC to other voltages DC is easy and inexpensive. Converting 110VAC or 240VAC to DC is not nearly so. I've been looking at converting the bulk of my home computers (which are all rack mounted) to 48V so that I can reduce the heat from the power supply and make things a bit more lean on the energy bill.

      --
      Do not look into laser with remaining eye.
    14. Re:Apple ... supercomputer...? by Anonymous Coward · · Score: 0

      the compartments are set up the way they are so they can get maximum airflow with minimal blowing

      That's what she said.

    15. Re:Apple ... supercomputer...? by Gilmoure · · Score: 1

      Not to mention the fact that these fans are low speed/high volume fans. They spin slower, producing less noise but moving more air than small, high speed fans.

      --
      I drank what? -- Socrates
    16. Re:Apple ... supercomputer...? by Phroggy · · Score: 1

      I have. Distributed power supplies means than one failure does not bring down the entire computer.

      I thought of this, but I assume it should be possible to have a few redundant power supplies all feeding into a common pool - obviously this is completely outside my field of expertise and my terminology is all wrong, but hopefully this makes sense. So, somebody who knows what the hell they're talking about, would this be feasible?

      I completely agree about some machines always being down, and eliminating single points of failure is obviously important for something like this.

      --
      $x='S24;r)>63/* h@<5+oZ)32"5cz';$me='phroggy'x$];
      $x=~y+ -xz+\0-Tx+;print$_^chop$me for split'',$x;
  10. Benchmarks? by grasshoppa · · Score: 0, Offtopic


    So, would this be like 2200 p4s running at 10ghz?
    </steve_jobs>

    --
    Mod me down with all of your hatred and your journey towards the dark side will be complete!
  11. Anyone have any real specs? by anzha · · Score: 4, Interesting

    So far we've seen that it's a cluster and what the building blocks are. What's the interconnect? What's the OS? What are the nodes using for a network filesystem? Are they at all? Is this intended for parallel jobs or for embarassingly parallel work?

    --
    Do you know why the road less traveled by is littered with the bones of the unwary?
    1. Re:Anyone have any real specs? by mfago · · Score: 4, Interesting

      The interconnect is Infiniband by Mellanox. These things get 10Gbps bandwidth with 6us latency under MPI. Very decent stuff. There is more information at the site above.

      Note that 1100*$3000 = $3M. This doesn't include the 4GB RAM, but also doesn't include any volume discounts. Thus the interconnect may cost about $2M.

      Oh, and to the guy who said "4 Athlons + Myranet is the same price as one G5" -- can I have some of what you're smoking?

    2. Re:Anyone have any real specs? by cosmo7 · · Score: 1

      Well, if you had 1100 dual G5s how would you cluster them?

    3. Re:Anyone have any real specs? by Kalak · · Score: 1

      OS is SuSE as it supports Infiniband.

      "Slow Down Cowboy"
      And now I'm typing to wait on /. to let me post this post that is something that might be relevant and on topic so others can post noise.

      --
      I am, and always will be, an idiot. Karma: Coma (mostly effected by .hack)
    4. Re:Anyone have any real specs? by JimRay · · Score: 4, Informative

      OS is SuSE as it supports Infiniband.

      Well, according to this story, the cluster will be running "a beta version of the latest release of OS X", presumably a beta version of Panther.

      If this is true, I'd bet, and this is purely a guess, that Panther and XCode, the new development tool built by Apple, have some support for cluster applications. With technologies like Rendezvous on top of Mach/BSD, it could mean beowulf style supercomputers that are both fast and easy to maintain.

      --
      My other computer is your Windows box
    5. Re:Anyone have any real specs? by mfago · · Score: 1

      Do you know they are using SuSE, or just guessing? I was wondering if Apple developed Infiniband drivers so they could tout OSX as the OS.

      I'd personally probably pick Black Lab on Yellow Dog Linux -- it's specifically developed for clustering, and includes libraries tuned for Altivec. Not that I don't like SuSE (I run it on my Thinkpad), but SuSE PPC is still at version 7.3, while Intel is at 8.2.

    6. Re:Anyone have any real specs? by rlowe69 · · Score: 1

      Note that 1100*$3000 = $3M. This doesn't include the 4GB RAM, but also doesn't include any volume discounts. Thus the interconnect may cost about $2M.

      There are, of course, other costs. The room(s) to house 1100 computers, air conditioning system, electricity, labor, maintainance and on and on ...

      --
      ----- rL
    7. Re:Anyone have any real specs? by GrumpyOldMan · · Score: 2, Informative

      To me, the Infiniband is the most interesting part. I'm very interested to see how the Infiniband scales and to see when they actually get the cluster working. I'm concerned about this because:

      1) I'm not aware of any Infiniband cluster that big. I'm not a big follower of IB, but I'd be surprised if there were any other clusters running MPI over IB even 1/2 that large.

      2) A Mellanox developer was asking basic questions about OSX driver development on the Darwin device drivers mailing list as recently as a few weeks ago. This leads me to believe that the MacOS X IB driver may not yet be ready for prime-time. Or may not even exist..

    8. Re:Anyone have any real specs? by Anonymous Coward · · Score: 0

      haha...what labor? they're begging for volunteers to put it together

    9. Re:Anyone have any real specs? by Kalak · · Score: 1

      OK moderators, read the story he just linked to. The only OS mentioned in it is mentioned by "Yankee Group senior analyst Dana Gardner" who has no stated connection to VT. I'm not saying that you should take my comments as correct (since I haven't bothered to track down a difinitve source on the OS, they are difinitive rumor at this point), but you can certainly check his source out.

      With technologies like Rendezvous on top of Mach/BSD, it could mean beowulf style supercomputers that are both fast and easy to maintain.

      I'm writing this from an OS X box, but that collection of words sounce like more marketing than anything else.

      --
      I am, and always will be, an idiot. Karma: Coma (mostly effected by .hack)
    10. Re:Anyone have any real specs? by JimRay · · Score: 2, Informative

      The only OS mentioned in it is mentioned by "Yankee Group senior analyst Dana Gardner" who has no stated connection to VT.

      From the article that I linked to:

      "In addition to the G5 machines, the university said it is using a beta version of the latest release of OS X, new networking hardware from Mellanox and Cisco, and cutting-edge configuration and cooling technologies to build the powerful cluster for a fraction of the price of a traditional supercomputer."

      (emphasis mine)

      Now, you can take that any way you like, I was simply trying to add another piece of information, which is why the post has been modded as informative. I realize this may have been easy to miss, it being in the second paragraph and all rather than being burried down at the bottom with your "Dana Gardner" tidbit, but there you have it. Re-read TFA then come back and complain.

      As for my "words sounce [sic] like marketing", well, that may be, but the fact is, automatic network configuration (which is exactly what Rendezvous is) would make 1100 clustered G5's easier to admin. And this flight of fancy of mine was based on the fact that Apple is already using Rendezvous-based clustering for XCode and Shake, their high end video compositing software.

      Flame on if you like, just make sure you got yer facts straight first, kid.

      --
      My other computer is your Windows box
    11. Re:Anyone have any real specs? by rlowe69 · · Score: 1

      Labor for everything else. The computer isn't going to be the entire labor cost of this project. Someone has to build the rooms to spec, install the air conditioning, blahblahblah ...

      --
      ----- rL
    12. Re:Anyone have any real specs? by g0at · · Score: 1
      I'd bet, and this is purely a guess, that Panther and XCode, the new development tool built by Apple, have some support for cluster applications.

      Why guess? Steve explained at the keynote, like it says right here:

      ...Distributed builds using Rendezvous, allowing your entire work group to take advantage of each other's machines to compile your projects...


      Or are you alluding to something else?
    13. Re:Anyone have any real specs? by Kalak · · Score: 1

      I could qualify for an unnamed person that is the university saying something, as I work for the university. A quote with no name is worth as much as an AC on /.

      A beta OS is what cought my attention, and I can't see anyone putting a cluter running a beta OS on something as visible as this. "The cluster has crashed" would not look good.

      To spare redundancy, I have a post in this thread about rumor, but form a better source above. This is one step away form an authoritative source (like one named in the the articles liked in the headline) that I won't name until I can talk to him. No sense in dragging someone's good name through my rumor mongering.

      --
      I am, and always will be, an idiot. Karma: Coma (mostly effected by .hack)
    14. Re:Anyone have any real specs? by BiggerIsBetter · · Score: 1

      So were talking about OpenMosix style clustering rather than Beowulf...

      --
      Forget thrust, drag, lift and weight. Airplanes fly because of money.
    15. Re:Anyone have any real specs? by spankalee · · Score: 1
      I've read some other articles that claim the cluster will be running Linux, but you bring up an interesteing point with XCode.

      XCode does have distributed compiling support. I don't know how common this is or isn't with Linux development tools, but with this type of cluster compile times would be incredibly reduced. Yet another reason an OS X cluster is attractive.

    16. Re:Anyone have any real specs? by Johnny+Mnemonic · · Score: 1


      A Mellanox developer was asking basic questions about OSX driver development on the Darwin device drivers mailing list as recently as a few weeks ago.

      post to the Darwin Device list, if interested. That was from July 14, so I hope they worked fast. Plus it doesn't look like they got any responses, at least in list. Maybe those in the know might be inclined to help out this worthy project?

      If the archives challenge for a pass, use "archives/archives" as the user name/pass pair.

      --

      --
      $tar -xvf .sig.tar
    17. Re:Anyone have any real specs? by Anonymous Coward · · Score: 0
      "OS is SuSE as it supports Infiniband.

      Well, according to this story, the cluster will be running "a beta version of the latest release of OS X","

      And this makes sense, because there is no active PPC/Mac development at SuSE. They only work on stuff for the IBM PPC based machines. I bet Panther is far more adjusted for the G5 boxes than LinuxPPC will be for a while. And so i guess it will run faster with GUI deactivated.

    18. Re:Anyone have any real specs? by Anonymous Coward · · Score: 0

      Amazing that you get a "+5 Interesting" for a post whose questions are frequently answered by reading the fucking article!!

  12. Read the previous /. article first by bodrell · · Score: 1

    There was a lot of inside info from people who work at Virginia Tech or go to school there. Lots of speculation and rumor, too, if you're into that sort of thing.

    --
    Si la vida me da palo, yo la voy a soportar Si la vida me da palo, yo la voy a espabilar
  13. enjoy your tuition increases kids by rtphokie · · Score: 0, Troll

    Hopefully they'll build a nice big window so all the students can see this beo^H^H^Hcluster of macs. Besure to leave the blinds up though because these same students will be working 2 part time jobs just to afford tuition.

    1. Re:enjoy your tuition increases kids by Aadain2001 · · Score: 0, Offtopic

      Why is this a troll? I think it is very relavent. Universities around the country have been hitting financial troubles, and they ALWAYS pass this off to the students in some form. Most students are there because of scholarships, part time jobs and financial aid. Very few are there completely on Mommy and Daddy's dollar. When a University builds something like this it is pretty cool, but who is paying for it? I doubt the administration will be taking a pay cut in order to pay for it. Maybe they'll just fire a few professors and put some more on research and grant writing full time, leaving fewer teachers for students. Or maybe they'll just charge every student more money under something like a "reasource fee". Point is, is it really financially responsible for the University to built this big computer?

      --
      Space for rent, inquire within
    2. Re:enjoy your tuition increases kids by ukyoCE · · Score: 4, Informative

      It is a troll because the money for the supercomputer came from a NSF grant for that specific purpose. Furthermore the university expects to make a five-fold return, as have most universities in the top-x supercomputers.

      Have you ever been to VT? We've got construction going on all over the place. The football stadium is about to get another "upgrade" after having received on just a year or two ago. We've got major construction going on in at least 3 different places, not to mention many smaller construction projects.

      Meanwhile teachers are getting let go, classes that were taught in 30-person rooms 3 years ago when I started, are now taught in 400+ person lecture halls.

      Does it suck? Certainly. However the money for the construction projects, football stadium, and supercomputer are all from grants, donations, and other means intended for a specific purpose. They can not legally take the money from a supercomputer grant or football stadium donation and use it to pay a teacher's salary.

      We have uneducated rants in the school paper at least once a week saying "why are we upgrading the football stadium if we cant pay teachers!@#$"

      Yeah, it does suck, but the university has no choice in the matter.

  14. what are they going to name it? by Anonymous Coward · · Score: 0, Funny

    hmm, what to name a mac-based supercomputer?

    deep-pink.

    1. Re:what are they going to name it? by dasmegabyte · · Score: 1

      Well, no. It'd be deep BONDI blue, if it were 1997 (hint: it's not). Nowadays, it's Deep Titanium or Deep Snow, something Virginia has never seen.

      --
      Hey freaks: now you're ju
    2. Re:what are they going to name it? by commodoresloat · · Score: 2, Funny

      Just call it the Big Apple.

    3. Re:what are they going to name it? by 47Ronin · · Score: 1

      Considering that these towers and every mac product in the current line is either silver-colored or white.. I don't see how pink factors into anything here. If colored computers are so "gay" then what do you think of makers like Alienware that build PCs which are yellow or green?

      --
      Those who laugh at you for you having a Mac.. are the people who constantly call you to fix their PC.
    4. Re:what are they going to name it? by calyphus · · Score: 1

      Make that Big Blue Apple.

      --


      The potato it is uninformed.
    5. Re:what are they going to name it? by bursch-X · · Score: 1

      How a bout Titan, ok it's made out of Aluminium...
      I got it:

      The Aluminati Cluster

      And I bet the first they're going to compute would be the question to "42".

      --
      There are two rules for success:
      1. Never tell everything you know.
  15. why g5 by mrtroy · · Score: 0

    support microsoft! do it with ten times as many xboxes!

    :) But seriously why would they specifically choose g5's....they must have done a bit of research before dropping 5 million on them

    --
    [I can picture a world without war, without hate. I can picture us attacking that world, because they'd never expect it]
    1. Re:why g5 by dema · · Score: 1

      i would guess they have a higher-up who is a mac fan that was probably pushing them, only because it's rare to see such a setup with macs.

    2. Re:why g5 by mrtroy · · Score: 1

      is it 5 million dollars worthwhile rareness?

      I would buy some rare *insert collector item here* and make a normal faster/cheaper/better cluster!!!

      --
      [I can picture a world without war, without hate. I can picture us attacking that world, because they'd never expect it]
    3. Re:why g5 by Theatetus · · Score: 1

      Eh, Virginia's tuition problems come from doing away with the car tag tax in the '90s, not building a supercomputer on the cheap in hte 'oughts.

      --
      All's true that is mistrusted
    4. Re:why g5 by weez75 · · Score: 1

      Contrary to popular belief, those building supercomputers don't just run down to the local CompUSA and pick out the one they like best. I'm quite sure that these folks did their homework and looked at the various options.

      Buying good hardware from a reputable company that commits to helping you meet your goals should never be shunned by a bunch of people who have never been faced with building one of the three fastest supercomputers in the educational world. If any of those people want to critique this purchase then I'm listening.

      --
      Of course we torture people, we need the information --Gen. Pinochet
    5. Re:why g5 by jceaser · · Score: 1

      Floating point operations.

      Why is it that no one is willing to give Apple an inch of credit?

    6. Re:why g5 by RevMike · · Score: 1
      why would they specifically choose g5's....

      Speaking very broadly, the recent PowerPC chips are very good at vector calculations, where the x86 chips tend to be better at integer calculations. Many scientific applications - for instance weather models - perform very well in vector models.

  16. Re:Virginia Tech by splatter · · Score: 1

    humm someone have a problem with rednecks?
    Maybe a little to close to home? Or is the poster from WVA?

    --
    "(I) have this unfortunate condition that causes me not to believe a single thing any politician says when a mic's on.
  17. Don'y let us find out where you live... by TWX · · Score: 1

    ... or else you'll be the victim of a drive-by fruiting!

    --
    Do not look into laser with remaining eye.
  18. 1100 G5s or PC processors ... by Anonymous Coward · · Score: 0

    ... is probably equivalent to 1 (well, maybe 2) IBM System/390s.

    1. Re:1100 G5s or PC processors ... by confused+one · · Score: 1
      the system/390 is based on power4 cpus, isn't it? The G5 is based on a power4 core. I don't see your logic.

      As another /. user pointed out, G5's do have an upgrade path. It's an IBM mainframe...

    2. Re:1100 G5s or PC processors ... by tmasssey · · Score: 4, Insightful
      This is a common mistake: confusing a supercomputer (or worse, a cluster) with a mainframe.

      Mainframes have one job: to move data from point A to point B as quickly as possible, while doing a relatively minor amount of processing on the way. Mainframes are what you use when you want to process every ATM transaction that happens around the world, all at the same time. In fact, your average mainframe is not really any more powerful than a dual- or quad-CPU Intel server, raw processing wise.

      Supercomputers are the exact opposite. They're stacks and stacks of CPU's that process largely independent chunks of data. They do huge amounts of processing on each chunk of data. They do *not* move data particularly well. In many cases, supercomputers are held together with Gigabit Ethernet. That's not exactly *fast*...

      Different computers, different tasks.

    3. Re:1100 G5s or PC processors ... by valdis · · Score: 1

      Please note that what IBM calls the G5 and G6 CPU cores used in the S/390 and Z-series are a DIFFERENT chip than the Power series. The big push in the S/390 series is I/O bandwidth and reliability. Even the smallest S/390 currently available will take up to 256 I/O channels - which will *each* support a number of Fiberchannel-class connections.

      Reliability - it does things like have 14 CPU dies on a card - 12 processors and 2 spares. Each CPU is actually 2 running in lockstep with a comparator. At the end of each instruction, the two sides are compared for "same state" - if they agree, the entire state of the processor is latched into a memory array. If they disagree, the hardware tries to clear a possible soft error by reloading the state from that array and retrying the failing instruction - if they agree this time, it logs a soft error and goes on. If they still disagree, another chipset called the 'processor controller' reads out that latch and loads it into one of the spares, and starts it running from there.....

      On the other hand, the 750 chipset that the world calls the G5 is a different beast, which will kick the S/390's butt on CPU speed. But it can't get anywhere near the z-series for reliability and I/O.

      http://www-1.ibm.com/servers/eserver/zseries/990 .h tml

    4. Re:1100 G5s or PC processors ... by confused+one · · Score: 1
      Ok, you got me. I didn't check to see specifically what a system 990 was. I was being a smart-ass and stretching the truth a little to make a point. I'm aware of the difference between a mainframe and a "supercomputer". I'm just trying to make a comparison for the idiots I'm replying to hoping they'll at least grasp the concept...

      I'm frustrated with people arbitrarily say the Power PC chips aren't capable, when I know they are (having worked on a cluster of PPC based VME processors doing processing and high I/O throughput to a central mainframe). Motorola just dropped the ball wrt the high-end market, etc. IBM designed the architecture and is putting the ball back into play so to speak.

  19. Obligitory Troll by Maskirovka · · Score: 5, Funny

    Slashdot summary:
    1) Itaniums are for pussies.
    2) Go Apple!
    3) Opterons still kick the G5's butt.
    4) I can't wait to run doom3 on my backordered G5.
    5) People griping about apples proprietary hardware and software, and how this cluster could have been built cheaper from oem parts, and ebay ethernet hubs.
    6)Dumb lists summarizing other trolls.

    1. Re:Obligitory Troll by Anonymous Coward · · Score: 0

      7)Replies to lists summarizing other trolls.

    2. Re:Obligitory Troll by Anonymous Coward · · Score: 0

      >6)Dumb lists summarizing other trolls.

      Shouldn't we should go recursive on his post and hence score it "-1, redundant"?

    3. Re:Obligitory Troll by EvilTwinSkippy · · Score: 0

      7) Obligatory "In soviet Russia the G5 compute YOU!"

      --
      "Learning is not compulsory... neither is survival."
      --Dr.W.Edwards Deming
    4. Re:Obligitory Troll by Anonymous Coward · · Score: 0

      People making lists and try to "prempt" the thoughts of other's is really fucking annoying.

      We don't need someone trying to summarize every article while at the same time pointing out how boring and predictable we all are.

      Parent: If you find the comments so repetitive and obvious then go fucking elsewhere.

      Unless of course you mods aren't tired of this hackish practice. Then by all means continue to mod the witless up.

    5. Re:Obligitory Troll by Anonymous Coward · · Score: 0

      Waaaaaaaaaaaaaaaaaaah!

      If you find his comments so offensive, then why don't *you* "go fucking elsewhere"? ;-)

    6. Re:Obligitory Troll by 47Ronin · · Score: 0, Offtopic

      7) ???
      8) Profit!!

      --
      Those who laugh at you for you having a Mac.. are the people who constantly call you to fix their PC.
    7. Re:Obligitory Troll by Anonymous Coward · · Score: 0

      Eat shit asshole. You and your lame reply can suck my dick.

    8. Re:Obligitory Troll by Anonymous Coward · · Score: 0

      Well spoken.

  20. Actually by Anonymous Coward · · Score: 0

    Poster went to William and Mary - which might suck even worse than VA Tech. Virginia is a police state, and its state colleges are underfunded havens for rednecks and trenchcoat mafia style rejects from NoVA.

    1. Re:Actually by Choobius+Gothicus · · Score: 1

      Considering this school is in the top 50 doctoral programs in Computer Science, it undeniably proves that you are more ignorant than the targets of your diatribe. This implementation will probably help their standing in the academic community, and is probably a good investment.

    2. Re:Actually by Anonymous Coward · · Score: 0

      i liek beef jerky

  21. Even more info ... by Pentagon13 · · Score: 4, Informative


    Here's the article from which the Collegiate Times article has paraphrased: http://www.technews.vt.edu/Archives/2003/Sept/0356 6.html

    1. Re:Even more info ... by n8_f · · Score: 1

      Very informative that. The best is at the end, where the Mellanox CEO says "[T]he combination of industry standard servers, Linux and InfiniBand...."
      So, they are going to run Linux on it.

      Although they mention "off-the-shelf" a couple of times, including prominently in the opening paragraph, it doesn't sound like the computers are what you get at retail. First, they are using PowerMacs and not Xserves, as noted here: "officials selected Apple's new Power Mac." However, it also says that "the university worked with Apple to purchase and adapt the new machines." I'm guessing this means they aren't using the retail case, which as many posters have pointed out would be insane for a cluster this size.
      The other interesting piece to note is that "Liebert...custom designed computer racks along with power distribution equipment", although they did use their "new high-density rack mounted cooling system" unmodified, it sounds like. Okay, the only way the G5s could approach a heat load for which "normal air conditioning units were insufficient" would be if they were out of their retail cases. You just can't achieve enough density otherwise. And while a custom designed rack could mean made to fit PowerMac retail cases, I think the likelier possibility is that they were significantly increasing density. A G5 has about half the heat output of an Athlon or P4, so they must be packing them in pretty dense.

      Anyway, the conclusion is that they are using a custom case and running Linux.

  22. Re:But WHY? by confused+one · · Score: 1

    Since it's based on a Power4 core, I think you should ask IBM that...

  23. Water main control by whipping_post · · Score: 0, Troll

    This will obviously be used for the hundreds of thousands of outhouses on and around campus.

  24. Re:Too expensive by EvilTwinSkippy · · Score: 2, Insightful
    G5: Deliverable today

    Opteron: Still under development.

    Now tell me, on the Good/Fast/Cheap curve you design parameters lie?

    --
    "Learning is not compulsory... neither is survival."
    --Dr.W.Edwards Deming
  25. Subsidized? by FatSean · · Score: 1

    How much would you bet that Apple is throwing in some $$ for this try.

    --
    Blar.
    1. Re:Subsidized? by Anonymous Coward · · Score: 0

      Apple doesn't play that way. Apple wins these kinds of things on technical superiority. Note Windows and Linux weren't even considered. So much for their much-vaunted "performance" claims. Once again, Apple is the de-facto standard for high performance computing.

  26. Re:PC Upgrade Woes by Anonymous Coward · · Score: 0

    I laugh every time I see this...truly a classic.

  27. Re:PC Upgrade Woes by orionware · · Score: 0, Redundant

    I just installed Acrobat 6 on an athlon 800Mhz we have here and it installed in about 80 seconds.

    You have either:

    a) a serious hardware issue with the P4 machine

    or

    b) a tendency to lie in an effort to boost the mac image

    Since you say you upgraded from a Cyrix M3/233 you might just be a simple moron since they haven't made that chip in God knows how many years and it was a shit architecture as well.

    The whole story is dubious as why would you buy a p4 serial ata to help you at your freelance gig where you "copy a 17 meg file from one folder on the hard drive to another folder"?

    --


    Karma means nothing to me, so suck it...
  28. Be nice to Tech by The+Ape+With+No+Name · · Score: 1

    It may be true that all dirts roads lead the VPI, but all white lines lead to UVa...

    --
    Comparing it to Windows will be a moot point, since El Dorado is going to have a 40% larger code base than XP.
  29. What the hell do you do with 1100 mice? by Eclypser · · Score: 5, Funny

    Post suggestions here!

    --
    The comment has already been made. Let's move it along people. Nothing to see here.
    1. Re:What the hell do you do with 1100 mice? by Anonymous Coward · · Score: 0

      Well, sell them on eBay, of course!

    2. Re:What the hell do you do with 1100 mice? by daeley · · Score: 5, Funny

      I don't know what you'd do with them, but I'm sure some wanker on here is already complaining there aren't 2200 buttons. ;)

      --
      I watched C-beams glitter in the dark near the Tannhauser gate.
    3. Re:What the hell do you do with 1100 mice? by Bob+The+Cowboy · · Score: 1

      build the world's largest 1100-button mouse?

    4. Re:What the hell do you do with 1100 mice? by Xugumad · · Score: 2, Funny

      Feed an army of robotic cats?

    5. Re:What the hell do you do with 1100 mice? by El · · Score: 1

      Take them apart and use the pieces to build 366 mice with the right number of buttons.

      --

      "Freedom means freedom for everybody" -- Dick Cheney

    6. Re:What the hell do you do with 1100 mice? by insanecarbonbasedlif · · Score: 1

      I could care less about the missing 1100 buttons, but where are my scroll wheels? I need to scroll.

      --
      Just because I doubt myself does not mean I find your position compelling.
    7. Re:What the hell do you do with 1100 mice? by dasmegabyte · · Score: 1

      Yes, lucky for him there are 2200 shift, alt, control and apple keys!

      --
      Hey freaks: now you're ju
    8. Re:What the hell do you do with 1100 mice? by Corporate+Drone · · Score: 1
      hmm... i'm not sure, but i'm thinking it was probably an interesting freshman orientation barbecue ...

      --
      mmm... yeah... You see, we're putting the cover sheets on all TPS reports now before they go out...
    9. Re:What the hell do you do with 1100 mice? by Anonymous Coward · · Score: 0

      that's the funniest damn thing I've ever read

    10. Re:What the hell do you do with 1100 mice? by hobbit · · Score: 1

      You use a 3.0054644808743169398907103825137-button mouse?!

      --
      "Wise men talk because they have something to say; fools, because they have to say something" - Plato
  30. Memory ? by hopbine · · Score: 1

    From one of the articles "For the supercomputer to break the top five supercomputers in the world, it would have to possess 10 teraflops of memory."

    --
    Semper ubi sub ubi
    1. Re:Memory ? by mwhahaha · · Score: 1

      I graduated from VT in May...The Collegiate Times is full of typos, gramatical errors and misprints. You have to take their news with a grain of salt. Silly wannabe journalists.

    2. Re:Memory ? by You're+All+Wrong · · Score: 1

      If it breaks the top 5 machines, I can see some pretty unpleasant lawsuits coming, and perhaps a wave of suicide bombing.

      YAW.

      --
      Your head of state is a corrupt weasel, I hope you're happy.
  31. The money came from a grant by jocknerd · · Score: 1

    The alumni have other causes to spend their money on, like renovations to the football stadium.

    1. Re:The money came from a grant by coolmacdude · · Score: 1

      No kidding. We (Georgia Tech) just completed phase 2 of our stadium renovation which cost 75 million.

      --

      -You may license this sig for only $6.99.
    2. Re:The money came from a grant by Anonymous Coward · · Score: 0

      Oh fuck don't get me started on the GaTech statium, meanwhile, they are talking about shuting down our HPC facility.

      I hate the athletic dep.

  32. It's obviously why Apple by orionware · · Score: 1, Insightful

    I would bet Apple would gladly cut a fantastic deal to get their model in the news as being part of one of the fasters supercomputers. They likely paid very little for the hardware.

    It's a no brainer...

    --


    Karma means nothing to me, so suck it...
  33. PowerMac G5s? by foo+fighter · · Score: 5, Interesting

    Why aren't they waiting for the Xserve update? Rhetorical question, but still...

    I haven't seen one, but it looks like the PowerMac G5s are about 4U wide. 1100 x 4U = 4400U / 42 per rack ~= 105 racks.

    Not only is this going to take up an enormous amount of room, but the power and cooling requirements are going to be crazy as well. And they don't have rails so getting them in the racks, and working on them once in the rack, is going to be a PITA.

    1100 G5 Xserves would need only about 25 racks. Many fewer UPSes and A/C units to power in each rack. Much easier to install and work on.

    I know Apple is gung-ho about this validating their "Fastest PC Ever" claims. But it seems a little poorly thought out on the University's part even if they got a sweet up-front price on the machines. Remember: the system price is a small part of TCO.

    --
    obviously no deficiencies vs. no obvious deficiencies
    1. Re:PowerMac G5s? by EvilTwinSkippy · · Score: 1

      When your are ordering 1100 of them, I'm pretty sure Apple with throw in the rack mount hardware. For all we know, they are getting a special production run with a custom case.

      --
      "Learning is not compulsory... neither is survival."
      --Dr.W.Edwards Deming
    2. Re:PowerMac G5s? by mfago · · Score: 1

      I agree that an Xserve G5 would be much more space efficient, but remember that VT supposedly went with Apple over HP/Dell because of ship date.

    3. Re:PowerMac G5s? by jchapman16 · · Score: 1

      The G5 is actually too tall to fit in a standard 19" rack horizontally, so only 6 G5 cases can fit in a standard rack on 3 shelves (case dimensions are 20.1" by 8.1").

    4. Re:PowerMac G5s? by hackstraw · · Score: 1

      I agree, thats one huge footprint. I used to work at Tech, and believe me, they have the space for such a system, but its still kinda wierd.

      Remember: the system price is a small part of TCO.

      Really. My system cost about $1mil and over 5 years it will cost less than $350,000 to run (cooling, electricity, rent, and me). Hell, after I pay my SCO licenses (NOT!) it would only be another $100k or so.

    5. Re:PowerMac G5s? by Pyrometer · · Score: 1
      hy aren't they waiting for the Xserve update? Rhetorical question, but still...

      There is a reason the G5 desktops have 9 fans in them ... and no it is not for the hard drives or for the Radeon 9800 Pro card in them ... it is for the G5! The IBM.G5 draws more power and is hotter than a G4 ... therefore having two G5's in the same XServe enclosure would probably not fit the bill ... unless you want a hot-plate.

      IF we were to see XServe machines based on the G5 I would expect 2U (possibly 3U) enclosures designed for them. Given that apple would want a new desktop before server means that obviously the first product based on the G5 would be a desktop. IF apple are planning on ramping up the server/enterprise market I would be expecting a new range of G5 XServe's coming for '04. Off course that is just wild speculation on my part based on no fact what-so-ever.

      If only I worked for IBM instead of moto maybe I would know ;)

    6. Re:PowerMac G5s? by godawful · · Score: 1

      i would assume when working out the deal with apple VT became privy so some knowledge about the release of a g5 xserve..
      and rather then wait (till whenever that may be, perhaps apple told them not for a while), they opted for the option available today (or 4 to 5 weeks as the case may be for some users)

      --
      Live EVERY week... Like it's Shark Week
    7. Re:PowerMac G5s? by jcr · · Score: 3, Insightful

      Why aren't they waiting for the Xserve update?

      What, there's an Xserve update?

      Take a look at the heat sinks in a G5. If you can figure out how to get that into a 1U enclosure, you might want to work for Apple in the hardware design group.

      -jcr

      --
      The only title of honor that a tyrant can grant is "Enemy of the State."
    8. Re:PowerMac G5s? by valdis · · Score: 4, Insightful

      To get it into 25 racks, you need to get it into a 1U form factor. At that point, cooling becomes an issue - you have 40 750W power supplies *per rack* then. We're talking about a space about the size of a phone booth, and the heat equivalent of 20 hair driers on HIGH all going at once. It's gonna get TOASTY at that point. Even if you consider a 2U and expanding out to 50 racks, that's still a lot of heat per square foot.

      Remember - the CPU and the memory are going to generate the same number of BTU/hour whether in a tower case or a 2U rack. And going to the *as yet unannounced* 2U just makes the cooling problem worse...

      Yes, we thought a LOT about these sort of issues.

    9. Re:PowerMac G5s? by fork420 · · Score: 5, Insightful

      Perhaps they have entered the Reality Distortion Field (TM).

      Apple's platform will probably work here, and once they prove it does, they will open a new segment of buyers, and gain plenty of positive attention. Apple's (recent) record suggests they will exceed expectations.

      As to the cost...the good PR that they can derive from a working G5 supercomputer is easily worth $10m on hardware. At the end of the quarter it's hardly noticable to a company sitting on $3,500,000,000.00 in cash.

      Apple gets to show MSFT they can scale OS X way past Windows. IBM gets to show Intel what it thinks of the Blue Lightning License arrangement, and Apple and IBM, by virtue of being together, get so give MSFT the finger...from way up the (top500.org) list. Not bad for their first try at this.

      ...and think about it from Jobs' perspective: This will be the biggest "Apple Computer" ever made. His ego can't handle this thing not kicking ass.

      --
      Don't forget that Pixar's always looking for high power hardware and they've got lots of new money.

    10. Re:PowerMac G5s? by in7ane · · Score: 1

      The way you put it... I'm just too excited!

      This should have been written on the front page!

    11. Re:PowerMac G5s? by deander2 · · Score: 0, Redundant

      these ARE G5 Xserves...just cuz the public can't get them yet, doesn't mean vt can't.

      btw, tech is a very geek-heavy/savvy campus. i wish i was still there. ;-p

    12. Re:PowerMac G5s? by BWJones · · Score: 1

      And going to the *as yet unannounced* 2U just makes the cooling problem worse...

      "He's gonna smoke a turd in Hell for that one."
      -Steve

      Hrmmm. :-)

      --
      Visit Jonesblog and say hello.
    13. Re:PowerMac G5s? by Anonymous Coward · · Score: 0

      There are going to be 192 PCs per row x 6 rows. I don't know how many racks are in a row. (I just got back from an information session about the new cluster.)

    14. Re:PowerMac G5s? by Anonymous Coward · · Score: 0

      Can be done. At my company we're preparing a system with 48 750Watters in a rack. Our cooling methods are a little unusual, but I'm sure that Apple can be creative in that area.

    15. Re:PowerMac G5s? by Hoser+McMoose · · Score: 1

      You can easily get a pair of G5's into a 1U server. Dell, HP and the like all ship 1U servers with dual 3.0GHz XeonDP chips in them, and those chips consume quite a bit more power than the G5. The G5 uses roughly the same amount of power as a similarly clocked AMD Opteron, and almost all of the dual-processor Opteron systems out there are 1U racks.

      The G5 uses 9 fans so that they could use relatively slow spinning and variable speed fans. By all accounts their systems are VERY quiet while idle, though not abnormally quiet once they spin up to full speed. It's a rather nice touch IMO, though I also went out of my way and spent a little bit of extra cash to make my AthlonXP system very quiet (ie I bought a big heatsink that cost $50 more than the small heatsinks with tornado-style fans that most users buy), and even then I ended up with a power supply fan that is a bit louder than my liking.

    16. Re:PowerMac G5s? by TheRaven64 · · Score: 1
      The 1.2GHz PPC 970 only dissipates around 20W of heat. It was designed for IBM's blades, and may well end up in PowerBooks in about 6 months, and will probably make it into X-Serves before then.

      It would be quite easy to fit one of these into a 1U enclosure (as long as the chipset isn't too power thirsty). Of course then you would have 1.2GHz of CPU power in a 1U enclosure, rather than 2x2GHz in a desktop case, so the CPU power / cubic foot would probably be around the same, or slightly lower. They would also have to wait for Apple to get around to doing this, which since they are on a tight schedule, may not be an option.

      G5 X-Serves will be around eventually, although for the SME server segment they are aimed at even a 1.2GHz G5 would be overkill.

      --
      I am TheRaven on Soylent News
    17. Re:PowerMac G5s? by Anonymous Coward · · Score: 0

      Did it ever occur to you that the rating on a power supply is not the power consumption but the maximum power output? A WinTel or Mac power supply, say 400 watt for an decent one is not going to be using up the equivalent as four 100 watt light bulbs simply because the 400 watt power supply can provide 400 watts while the light bulbs actually consume 400 watts.

      Now for shits and giggles to make your example actually true, let's see what the normal G5's power supply would be supplying for .....
      1) Since G5's take less energy than P4s, using a P4's power rating would be in your favor. 100 watts. Make that 2 processors, so 200 watts.
      2) Memory/board/Video Card Another 50 watts. Yeah, I pulled that out of my ass, but it sure sounds reasonable technically.
      3) Monitor. None. (remember, Apple G5's power supplies also supply for ADC monitors, including CRTs)
      4) CD Rom Drive (5 watts)

      If a G5 has a 750 watt power supply like you said (which I'm not sure, I would have guessed 450 at most), this leaves a little under 450 watts for harddrives. A 7200 rpm harddrive is what? 20 watts on startup, 15 constant? Let's go for 22 harddrives in that G5.

      Sure, I'd believe a dual 2ghz G5 would be like a hair dryer if we had 22 harddrives in it. Yeah, I think it'd be pretty toasty if we had 22 * 1100 harddrives in it too. Full of hot air, just like your calculations.

  34. Their Website by rabbit994 · · Score: 2, Interesting

    Pudge thinks their website isn't good enough. What does he want? Some flash? Maybe some pop up ads to spice it up. Whatever happen to simple being good and fancy being woooo pretty but useless. Oh wait, that still hasn't changed.

    1. Re:Their Website by Anonymous Coward · · Score: 0

      Yeah really. The website is fine. The search is useless because it pulls up all these pdfs and out-of-date webpages, but the main page works fine. Not only that but it is nearly identical to most university's webpages, and for good reason. It's an easy to use low-tech page.

      Pudge can go fuck himself. As if his webpage is any better.

  35. Can you imagine.. by scumbucket · · Score: 0

    a Beowulf cluster of..... oops sorry!

    --
    CMDRTACO CHECK YOUR EMAIL!
  36. Not really a dupe by Kalak · · Score: 2, Interesting

    This isn't really a dupe, as this is a mention of the first official words form the school on the subject. Officials are finally speaking (and in some cases backing off) of the cluster in public.

    --
    I am, and always will be, an idiot. Karma: Coma (mostly effected by .hack)
  37. Imagine a.... by Joe+Decker · · Score: 1
    Imagine a cluster-computer post where nobody asks me to imagine a ....

    Oh, nevermind.

    1. Re:Imagine a.... by caluml · · Score: 1

      And as if by magic, 1 minute earlier....

  38. Floppy memory?? by Stephen+Samuel · · Score: 2, Insightful
    From the second article:
    For the supercomputer to break the top five supercomputers in the world, it would have to possess 10 teraflops of memory.

    I think that they mean 10teraflops of computing power, as opposed to 10terabytes of memory -- since the later would require each CPU to have 10GB of ram in it. Nonetheless, the anomaly tells me that this is a reporter not used to computer issues. (too few computer geeks at the college paper).

    --
    Free Software: Like love, it grows best when given away.
    1. Re:Floppy memory?? by Anonymous Coward · · Score: 0

      Heh, this paper hardly has a clue on any issues, much less computing. If the Collegiate times isnt one of the worst college papers around then I feel sorry for those schools with worse papers.

    2. Re:Floppy memory?? by Anonymous Coward · · Score: 0

      Interestingly enough they have now changed that line on the papers website to read

      "For the supercomputer to break the top five supercomputers in the world, it would have to possess 10 teraflops of performance."

      Rather than printing a retraction we might as well just pretend it didn't happen. Well the print copy I have still shows thier lack of understanding.

    3. Re:Floppy memory?? by Anonymous Coward · · Score: 0

      thier lack of understanding

      Similar to your lack of understanding of basic spelling?

    4. Re:Floppy memory?? by Anonymous Coward · · Score: 0

      A typo isn't quite the same thing

    5. Re:Floppy memory?? by Space+Coyote · · Score: 1

      Don't you see? Floppy memory is brilliant, they're taking advantage of the cheapest, most readily available storage medium in existence: Trillions and trillions of AOL Floppy disks.

      --
      ___
      Cogito cogito, ergo cogito sum.
  39. And the next day... by Anonymous Coward · · Score: 0


    And the next day, Apple lowered the price of the G5 by $150, and doubled the RAM, HD space, and included DVD-recorders.

    1. Re:And the next day... by Anonymous Coward · · Score: 0

      All G5 systems include DVD recorders.

      Dumbass.

  40. Ever changing focus shift.. by Rinikusu · · Score: 5, Insightful

    Anyone get the feeling that Apple might be pulling a Be, Inc and is trying to pull off a focus-shift?

    Remember Be, the "multimedia" OS turned "Internet Appliance". Remember the death of Be. (damn, that stings. I miss the BeOS.)

    Now witness Apple:
    For decades, seemingly the darling of the press-production (DTP) world, catering to artists of all magnitudes, it was the computer you used to create real, bona-fide art. It attracted the freaks, the hippies, the art chicks. For many people, this was unnerving. Different people get "different" looks.
    Now who's Apple targetting?
    With OS X, I'm thinking geeks. We're different people, too, but in a, well, different manner. Instead of the artists, Apple's going for traditional suits, the realm of IT. It may be a matter of sheer survival that Apple penetrates here, because they don't stand a chance in these days of "homogenous" work environments.. Out with Apple (even if it works) and in with Dell WinXP machines! Linux faces the same dilemna, although Linux has some other benefits/detriments for it's widespread adoption. If Apple can show it's worth in the server room (just like Linux is doing), then maybe, just maybe, they'll start looking at Apple on the corporate desktop (just like Linux is doing).
    Now, the idea of catering to suits is somewhat.. frightening. The whole damn market is different. They don't care about "look and feel", they care about numbers (see economic downturn, outsourcing to India, massive layoffs, H1B abuse, etc). This means Apple will have to change from being "cool" to utilitarian. But wait, I think I just painted myself into a corner here... Wasn't that the point of Apple? To be a tool and not an obstacle? Instead of creating computer art, we're now creating databases? Maybe Apple is on to something here...

    --
    If you were me, you'd be good lookin'. - six string samurai
    1. Re:Ever changing focus shift.. by the+idoru · · Score: 1

      yeah, being cool and being utilitarian aren't mutually exclusive. i have a 12" powerbook. best laptop i've ever owned for a number of reasons. yeah it looks super cool and the OS is really slick, but more than that it's an amazing tool. i truck it everywhere. the thing is very sturdy, rarely crashes, all the ports are in the same damn place, etc. bottom line is that it has increased my productivity as far as laptop-based work goes. artsy people might use Apples in part because they look cool, but when it comes to creating their works, i think they want the best tool for the job too.

      Apple has pushed its "it just works" mantra for a while now. maybe they are trying to transition that into the IT sphere where, in part because of the shitty economy and drum-tight budgets, the suits are beginning to look at things like computer downtime very seriously. Apple's argument would probably sound like: hey our computers rarely crash, rarely break, don't get those nasty virii, and on top of all that your workplace would look pretty damn cool with one on every desk. oh and we don't have those restrictive liscensing schemes. TCO is a term that's getting bandied about quite a bit. many businesses are willing to pay a little more up front for the hardware if the lifetime TCO is cheaper. i think they have to start out with small businesses where the money really is tight and there aren't a bunch of PCs already sitting around that will worry them about compatibility (as big a misconception as that is).

    2. Re:Ever changing focus shift.. by dasmegabyte · · Score: 1

      Problem with your post is it's way too speculative.

      First off, with OSX Apple isn't appealing to geeks...they're adjusting the geek's tools to a more public audience. Finally having a CLI does not change my graphic design friends' lives at all. Having a stable OS does.

      Apple's marketting hasn't changed much. They're still focusing on artists and home users. Front page of Apple.com routinely shows the box a computer comes in, or some software for making movies or music. Whereas Dell usually shows a multiethnic office with some guy pointing at a flat screen. Artful simplicity vs. complacent productivity. No change there.

      All Apple is doing is trying to rise up from being the whipping boy of the hardware industry. To show that they can compete where "normal" computers compete. To show that they're still making computers, and that those computers can crunch numbers that don't apply to photoshop filters. It's not a paradigm shift. It's cleanup PR. Apple knows damn well that the prospect of switching a whole office to a proprietary platform is scary, no matter how good it is. They'll make it easy, they'll reassure the switchers...but they're not going to break the bank marketting on the lost cause of the corporate IT department.

      And the market? It's not about numbers. It's about GREED. It's about executives getting million dollar raises for saving $500k. That math will only work so long. Apple's way has always been effective, it's still just as effective, the numbers are just much larger. The x86 PC market will see its margin slip to the point that new PCs will have to be nearly unusable to be cost effective. And that's when Apple will shine.

      --
      Hey freaks: now you're ju
    3. Re:Ever changing focus shift.. by jceaser · · Score: 1

      Very well writen. I total agree.

    4. Re:Ever changing focus shift.. by Anonymous Coward · · Score: 0

      Well,

      I'm not too sure this is much of a shift for Apple.

      From time to time Apple's always tried to lure in cash from IT and "the suits". It's just that in the past they didn't really have the right tools to make it happen; that and mis-management of these efforts always seemed to cause problems.

      I think Apple is casting a wider net these days, more than shifting focus. A tough ballence to be sure. How to gain "the suits" attention to the Unixy Goodness of MacOS X without alienating their traditional Educational and Artistic user base. But it is neccessary that they try to win hearts and minds in IT, if only to fight for the educational market filled by ITs pushing Macs out in droves for Dells.

      I want to believe they'll get it right this time, for whatever thats worth.

    5. Re:Ever changing focus shift.. by agent+dero · · Score: 1

      The difference here is that Apple can do both, that's their greatest victory with MacOS X.

      Unix for the geekoids, and the great MacOS for those artist folk.

      Anyways, a power processor with an Altivec-like twist, who cares! It does both.

      And, believe it or not, no company has ever been successful marketing a computer towards one type of user, (artist, geek, etc) It's about flexibility
      (P.S. I've been using Mac for web design, and coding since System 7)

      --
      Error 407 - No creative sig found
  41. Are they buying the chips from Apple or IBM by hshana · · Score: 1

    because I know that Georgia is buying a 256 node cluster and at least some of them will be G5's from IBM running linux.

    1. Re:Are they buying the chips from Apple or IBM by mrgoatCEO · · Score: 0

      It's not likely that they would purchase G5's from Apple; they don't deal in the OEM chip market, and their G5 towers are too bloated (expensive vid cards, etc.) for Supercomputer applications. IBM is the place to get the chips and applicable hardware, and the cluster itself will surely be linux based.

      --
      --Goat
      CEO, Goat Software
      Goatblog
    2. Re:Are they buying the chips from Apple or IBM by TimeZone · · Score: 2, Informative
      AFAIK, IBM does not have any systems based on the G5 (aka the PowerPC 970). Only Apple is actually making systems out of these chips. If somebody says they're using G5s, they're almost certainly using Apples. Or, they're confused and mean they're using some IBM Power5 box. Confusingly, the G5 is not based on IBM's Power5, but the Power4.

      TimeZone

  42. woohoo! by Anonymous Coward · · Score: 0

    well at least my $56k in tuition went to some good

  43. Wayback machine by pjdepasq · · Score: 1

    This was also discussed even earlier.... on Slashdot in July in a discussion of grid computing....

  44. Argument for G5 here. by eddy · · Score: 5, Interesting

    I'd love to see the arguments for the different platforms!

    I think the argument for G5 came from here.

    --
    Belief is the currency of delusion.
    1. Re:Argument for G5 here. by Anonymous Coward · · Score: 1, Funny

      Do one of those come with every G5?
      If so I'm switching platforms!!

    2. Re:Argument for G5 here. by Anonymous Coward · · Score: 0

      someone mod the parent funny because it's hilarious!

    3. Re:Argument for G5 here. by Anonymous Coward · · Score: 0

      No wonder Apple computers are expensive !!

    4. Re:Argument for G5 here. by Anonymous Coward · · Score: 0

      Man I'd like to fuck her.

    5. Re:Argument for G5 here. by Anonymous Coward · · Score: 0

      But you never will, will you? Sucks to be you...

    6. Re:Argument for G5 here. by bobrk · · Score: 1

      Obviously, Mac users have the hottest wives.

  45. Re:Well, by mrtroy · · Score: 0, Offtopic

    hahaha when has the answer to life NOT been told to us by the simpsons?

    --
    [I can picture a world without war, without hate. I can picture us attacking that world, because they'd never expect it]
  46. maybe not as cheap as you think by Anonymous Coward · · Score: 1, Informative

    Just the motherboard and chip for a dual 1.6 GHz opteron runs you $1394.99 if you need PCI-X (which is probably necessary with the interconnect you want). And on top of that, you need the network cards, memory, case, hard drives, etc. etc.

    You could easily get to Apple prices going with AMD.

  47. I'm sure Apple DONATED or SUBSIDIZED by Anonymous Coward · · Score: 0

    these machines for the publicity.

    That's the only explanation that makes sense.

  48. "Beef up their Website" by suwain_2 · · Score: 4, Insightful

    Can someone explain the " Maybe VT can use the new computers to beef up their web site" comment? It loads perfectly fast for me. It looks pretty good. It even runs PHP, so it couldn't be a "They shouldn't use ColdFusion" type remark.

    Am I missing something, or was that just a completely random comment?

    --
    ________________________________________________
    suwain_2 :: quality slashdot p
    1. Re:"Beef up their Website" by Anonymous Coward · · Score: 0

      It was a random, inane, and completely useless comment. This, coming from someone with a joke for a website. If VT's website needs beefing up, pudge's website is the cow that died 5 years ago and has since been picked clean.

      Seems to me the design for Virginia Tech is just fine. They could stand to get rid of their table-based layout, but it's not a horrible site. Pudge is just being an asshole.

    2. Re:"Beef up their Website" by shemnon · · Score: 1

      Apparently they already have the cluster up and running and it has since started to host their web site since the story was first posted.

      --
      --Shemnon
    3. Re:"Beef up their Website" by papadiablo · · Score: 1

      I think the comment may be referring to the Virginia Tech Corporate Research Center, which does appear to load things a bit slowly and does appear to be using ColdFusion for some things.

    4. Re:"Beef up their Website" by zerocool^ · · Score: 4, Informative

      It had better run freaking fast. Virginia Tech has had an OC-3 for at least 6 years, and I think they're upgrading to hook into network virginia's bigger pipes.

      For those out of the loop, network virginia is a partnership between verizon (local loops), sprint (borders and pipes), and Virginia tech (expertise and tech support). A few years ago, they had 2 OC-3's from Northern Va to roanoke, 1 to richmond, and 1 from roanoke to richmond. Their updated network topology map can be found by clicking here. The bottom one is the latest one. At any rate, they've got multiple OC-12's running from Nova to Roanoke, mainly because of VT. Tech may already be hooked into the OC-12's, i'm not sure.

      Also, I'm not sure about how much will be lost in clustering, but according to the CT article today, the dual 2.0 Ghz G-5 can pull 14 teraflops by it's self. If we're getting 1100 of them, say, drop ~10% for overhead, that would still put us up at 14000 teraflops, which is ahead of ascii white and behind los alamos.

      Also: regarding power requirements and all of that - we have several state of the art facilities on campus for this kind of stuff, including the VT Corporate research center and Torgersen hall (home of the center for advanced computing and where we keep all the fun VR rooms and stuff). There's a power plant on campus. We never lost power when I lived in a dorm, not during snow storms or huge thunderstorms or anything. It supplies power for most of blacksburg, too. Shameless plug, but that's one selling point for the company where I work, netmar, because we get our power from the VT power plant, and it's about 2.5 blocks away, we hardly ever lose power for more than 2 minutes, so we haven't had to put our generators to work in forever. Nowadays, we just test them with the remote start to make sure they're working, and to scare people that are hanging around the generator hut.

      Anyway, VT has no problems finding a place for these things to go, and will have no problem providing power for them. Climate control should be no problem, either. For starters, it's easy to cool things in blacksburg, cause it hasn't been above 100 degrees in 100 years here.

      Some people in my econ class today were talking about why are we doing it, and what's it going to be used for. Really, I think we're doing it to get grant money and sponsorships/funding, because with the economic situation in VA, we're scrambling to find money. We've had to drop teachers without replacing them and cut back on services all over (no more trash cans in dorm hallways, you have to take your own trash outside, can't afford the maintinance staff). Also, the Vet school will get a lot of use out of it. That's the "virginia-maryland regional college of veteranary medicine". They're looking for ways to cure problems with small bacteria instead of drugs (i'm not clear on the particulars, that's the impression i got). They're going to try and track what happens to something when it's introduced into an animal or something. Anyway, they'll use it, as will VT's engineering school, which, despite being tied for like 73rd on the list of top schools, and inexplicably 55 positions behind UVA, is an excellent program and produces excellent engineers.

      ~Will

      --
      sig?
    5. Re:"Beef up their Website" by Anonymous Coward · · Score: 0

      Can someone explain the " Maybe VT can use the new computers to beef up their web site" comment?


      Maybe it was anticipating the inevitable Slashdotting.
    6. Re:"Beef up their Website" by AllMightyPaul · · Score: 1

      Tech got an OC-12 this year directly to campus.

  49. You should watch out by jbarket · · Score: 4, Funny

    Jokes like this can get you put away in the punatentury for a very, very long time.

    --

    -----
    jonathan barket
    1. Re:You should watch out by Anonymous Coward · · Score: 0

      I guess that would be an example of punishment.

    2. Re:You should watch out by babbage · · Score: 1

      You asked for it... :-)

      pun moratorium, n.
      The doomed campaign to deoxymoronize computer humor.

      In particular, the vain attempt to demonstrate that plays on words such as RISC and UNIX are unfunny if the player is unaware of their historically built-in playfulness. Other puns assinorum deserving a well-earned retirement relate PARADIGM, Paradise, and ten sents in boringly obvious ways: "Paradigms Lost and Regained," "Brother, can you s'paradigm?" and so on. Likewise, the cash/CACHE thing is surely bankrupt: "Cache-only memory, no checks." Be assured, too, that every known C homophone has had its weary, C-sick day at the C-Users Journal's annual C-pun contest: C-through UNIX, C'est C Bon, Holy C, Proficient C, Vitamin C, O say can you C? e = mC^2, Variations in C, C-C Rider, Rauchen C?, The Cruel C from Cmantec, Mer-C Beaucoup... ad nau-C-am.

      One of Western Democracy's major flaws is that we cannot, without pettifogging legal interference, publically hang, draw, and quarter C-punsters.

      Quoted from The Computer Contradictionary, by Stan Kelly-Bootle, 1995

    3. Re:You should watch out by The+Munger · · Score: 1

      The vice of punning is a terrible vice indeed. There's no vice versa.

      --
      Refuse to make a statement in your sig!
  50. Pfft. by Anonymous Coward · · Score: 0

    Lies. Opteron boxes has been shipping for some time now.

  51. I agree by ArmenTanzarian · · Score: 1

    I just graduated from there and that website, while not ugly, is non-sensically arranged and the search comes up with some of the most ridiculous links. The department does have some nice clusters already set up though as well as a sweet 3d visualization studio!

  52. Re:But WHY? by Anonymous Coward · · Score: 0
    Since it's based on a Power4 core, I think you should ask IBM that...

    it's not

  53. Sounds fine by Doesn't_Comment_Code · · Score: 3, Insightful

    There are a bunch of people posting gripes that this was a bad idea. But I don't think it's that bad. We should at least withold judgment until we see some data. One thing's for certain, it will outperform YOUR cluster.

    Among the top complaints were:

    You could buy several AMD's for that.
    You might be able to, but the G5's they are buying already have 2 very good processors. As long as they're dividing up tasks among processors, it's nice to have all the memory management and overhead taken care of at a level of two processors per node instead of one. To be honest, I've never seen it done before, and it could have very interesting results.

    The Mac's aren't designed for this sort of thing.
    We don't know all the details of this cluster because they weren't all mentioned in the story, but my hunch is that Apple might cater to them a little if they are offering to dump $5 mill on a cluster. They might package the cases differently (sans curvy plastic or with shared power supplies).

    Anyway, when it comes to speed of high precision calculations, the G* chips have proven their worth. And most High Science applications fall into that range of operation. We all know that clustering and distrubuting is touchy. The cost and speed don't scale linearly. And the cost vs speed ratio definately doesn't scale literally.

    There is a possibility these computer science professors know something. So we might want to see how this thing performs before we rush to judgement.

    --

    Slashdot Syndrome: the sudden, extreme urge to correct someone in order to validate one's self.
    1. Re:Sounds fine by Anonymous Coward · · Score: 0
      A metacomplaint, about "AMD's" and "Mac's":

      You shouldn't add an apostrophe to a noun just because it's plural.
      The apostrophe has several proper uses in English, including indication of possession and indication of a contraction. It is not proper to insert one into plural nouns before the "s" for no reason.

      I'd also point out there's a possibility that the people with PhDs at other institutions who recently ordered Itanium2 or Opteron clusters also knew something. (This something might be that different jobs have different 'optimal' computer designs... or it might be that someone made non-optimal choices, or that they were offered different deals from manufacturers.)

    2. Re:Sounds fine by penguin7of9 · · Score: 1

      You could buy several AMD's for that.
      You might be able to, but the G5's they are buying already have 2 very good processors. As long as they're dividing up tasks among processors, it's nice to have all the memory management and overhead taken care of at a level of two processors per node instead of one. To be honest, I've never seen it done before, and it could have very interesting results.


      Dual processor AMD Opteron systems are 64bit, performance-competitive with G5, and considerably cheaper. Furthermore, you can choose among half a dozen vendors, and you can get them in rack mounts.

      Anyway, when it comes to speed of high precision calculations, the G* chips have proven their worth.

      That's a myth Apple has created by harping on rarely achievable AltiVec performance figures. And people who are happy with their Macs for desktop use like to give lip-service to that myth because it gives them a better excuse to buy Macs for servers and compute clusters.

      In real life, the G4 is a couple of years out of date in terms of performance, and the G5 is basically just keeping up with Opteron and Itanium.

      Apple's machines aren't bad, in particular for home and office use, but they are not the super UNIX workstations and scientific powerhouses Apple claims.

    3. Re:Sounds fine by RyuuzakiTetsuya · · Score: 1

      This sounds fine and dandy, however, where's your proof?

      I'm no Apple Zealot, hell, I'm something of an apple bigot. But I'd like to see numbers that back up my prejudice.

      --
      Non impediti ratione cogitationus.
    4. Re:Sounds fine by Anonymous Coward · · Score: 0

      This sounds fine and dandy, however, where's your proof?

      Well, since Apple is touting specFP now, I'd suggest you have a look at the score of the 1.5GHz Itanium2. (Hint: It's almost twice that of the G5).

    5. Re:Sounds fine by Gogo+Dodo · · Score: 1

      The original poster was comparing AMD processors, but okay... Itanium 2 is faster at specFP. However, where can you get a dual-processor 1.5Ghz Itanium 2 unit for $3000?

  54. Re:Too expensive by Anonymous Coward · · Score: 0

    The store I work at has Opteron's in stock. How does product in hand = still in development?

  55. "speculation and rumor"?? On /.?!? by sczimme · · Score: 4, Funny


    Lots of speculation and rumor, too, if you're into that sort of thing.

    This is slashdot! We're all about speculation and rumor. Innuendo, too, especially on the weekends.

    Oh, and sentence fragments.

    :-)

    --
    I want to drag this out as long as possible. Bring me my protractor.
  56. I keep overestimating slashdot... by Lally+Singh · · Score: 4, Informative

    Lots of "WHY?" questions, with lots of pointless trolling on the G5; but none of them actually look for answers. Mostly just more idiots who can't understand that a good vendor is important; that their own time is important; that ease of use is even more important now than it ever has been before. Luckily, these same idiots spend all their time setting up sendmail over their 14.4 modem. As for the G5, here are some strongpoints for it: - A fast memory pipe (1GHz) - Good heat management (9 fans but it's quieter than its predecessor) - Damn good FP performance (To get comparable FP performance on intel, you have to use the -fviolate-ieee flag on gcc, think about that) - Vendor-installed, vendor-supported Unix, with the vendor employing the entire OS's development team. - Fast system interconnects with network & I/O - Easy system setup (this matters a lot when you've got 1100 of them) - Proven apple reliability (and if you're going to fight this one, have something better than "is not!") (again, very important when you've got 1100 of them) Oh yeah, and OS X. Mach microkernel, Rondezvous, and distributed builds in the default toolset. Again, the idiots I mentioned above wouldn't have a clue about this stuff. As for _why_ VT getting this, VT's one of the largest engineering schools in the country. We've gotta simulate airflow over wings, heat propogation over materials, and other stufff this CS major doesn't understand. And we've got big development in bioinformatics. All kinds of CPU to crunch. AFAIK, the cluster's being paid for by federal grants or something like that. And now fools, flame me. Prove me right.

    --
    Care about electronic freedom? Consider donating to the EFF!
    1. Re:I keep overestimating slashdot... by Lally+Singh · · Score: 2, Interesting

      (and of course, after all that intellectual trash-talk, I posted without preview. here it is again, with the right (semi-) formatting). Lots of "WHY?" questions, with lots of pointless trolling on the G5; but none of them actually look for answers. Mostly just more idiots who can't understand that a good vendor is important; that their own time is important; that ease of use is even more important now than it ever has been before. Luckily, these same idiots spend all their time setting up sendmail over their 14.4 modem. As for the G5, here are some strongpoints for it: - A fast memory pipe (1GHz) - Good heat management (9 fans but it's quieter than its predecessor) - Damn good FP performance (To get comparable FP performance on intel, you have to use the -fviolate-ieee flag on gcc, think about that) - Vendor-installed, vendor-supported Unix, with the vendor employing the entire OS's development team. - Fast system interconnects with network & I/O - Easy system setup (this matters a lot when you've got 1100 of them) - Proven apple reliability (and if you're going to fight this one, have something better than "is not!") (again, very important when you've got 1100 of them) Oh yeah, and OS X. Mach microkernel, Rondezvous, and distributed builds in the default toolset. Again, the idiots I mentioned above wouldn't have a clue about this stuff. As for _why_ VT getting this, VT's one of the largest engineering schools in the country. We've gotta simulate airflow over wings, heat propogation over materials, and other stufff this CS major doesn't understand. And we've got big development in bioinformatics. All kinds of CPU to crunch. AFAIK, the cluster's being paid for by federal grants or something like that. And now fools, flame me. Prove me right.

      --
      Care about electronic freedom? Consider donating to the EFF!
    2. Re:I keep overestimating slashdot... by Anonymous Coward · · Score: 0

      > Vendor-installed, vendor-supported Unix, with the vendor
      > employing the entire OS's development team.

      If you want BETTER vendor-installed, vendor-supported unix wth the vendor mploying the entire OS's development team, you should go Sun/Solaris

    3. Re:I keep overestimating slashdot... by stratjakt · · Score: 1

      And whats missing?

      OS/X is out the window, the cluster will run SuSe.

      And I, for one, cant see choosing an architecture that does not support ECC memory for scientific applications.

      Yay, you ran a test across 1200 CPUs for a month - and your answer could have spontaneously become useless because someone walked into the lab with wool socks or there was a solar flare and a bit flipped on some box.

      Apple/IBM greased the wheels to some extent to get the G5s in there. This isnt shocking or surprising. (Well, it was the other day when /. reported on MSFT trying to increase its presense on campus)

      --
      I don't need no instructions to know how to rock!!!!
    4. Re:I keep overestimating slashdot... by Anonymous Coward · · Score: 0

      No way, the cluster will be running OS X. Why would you buy kick ass hardware then put a hobbyist operating system on it that doesn't even run Photoshop?

    5. Re:I keep overestimating slashdot... by ananke · · Score: 1

      and where exactly did you hear about suse? last time i spoke in person with folks who are setting this thing up, no word of suse was mentioned.

      --
      --- d'oh
    6. Re:I keep overestimating slashdot... by Erich · · Score: 1, Interesting
      There has got to be some sort of random apple troll comment generator.

      Just follow these easy steps:

      One: Baseless flame against everyone who disagrees with you:

      Mostly just more idiots who can't understand that a good vendor is important; that their own time is important; that ease of use is even more important now than it ever has been before. Luckily, these same idiots spend all their time setting up sendmail over their 14.4 modem.

      Two: Copy random specs from Apple's web page:

      A fast memory pipe (1GHz) - Good heat management (9 fans but it's quieter than its predecessor) - Damn good FP performance

      Three: Straight-out lies and made up stuff:

      To get comparable FP performance on intel, you have to use the -fviolate-ieee flag on gcc, think about that [...] Proven apple reliability

      Four: more flaming everyone else:

      Again, the idiots I mentioned above wouldn't have a clue about this stuff.

      Five: Claim to be superior:

      We've gotta simulate airflow over wings, heat propogation over materials, and other stufff this CS major doesn't understand. [...] All kinds of CPU to crunch.

      That's all it takes!

      Sigh.

      I haven't done real system administration for quite a while, but it's still blatantly obvious that you've never really had to deal with the administration of a compute cluster.

      Apple doesn't have a proven reliability record. At least, not in the enterprise server arena. Sun Enterprise Servers do. HP does. Apple? No.

      How can you seriously consider something like this without ECC memory? Do you really think that running 1100 copies of MacOSX on 1100 hard drives is a reasonable way to run a compute cluster? Do you have any idea how unreliable that would be? Do you really believe that Apple's Rondezvous will get everything setup perfectly for Infiniband?

      I see. You really do. How unfortunate. If you really are from VT, you represent it poorly.

      --

      -- Erich

      Slashdot reader since 1997

    7. Re:I keep overestimating slashdot... by tesmako · · Score: 1

      As has been mentioned in earlier posts it appears a great part of the purpose of this computer will be to test out their software system for fault tolerance. Probably interesting research.

    8. Re:I keep overestimating slashdot... by overunderunderdone · · Score: 1

      Do you really think that running 1100 copies of MacOSX on 1100 hard drives is a reasonable way to run a compute cluster?

      Umm... No. I'd imagine they would be usingNetBoot. It is built right in to MacOS X.

    9. Re:I keep overestimating slashdot... by Lally+Singh · · Score: 1

      Enterprise server arena? You do realize that the competition being listed here on /. is joe-blow's homemade athlon box x 1100 units. And ECC Memory? Again, compare to joe-blow. This is all about price-performance.

      As for proven reliability. Compare prices on ebay. The mac lasts longer.

      And when is saying that the vendor, my time, or ease of use are important baseless???

      Ah yes, the traditional "I'm the superior troll formula:"
      1. Don't say you're on top of stuff, but claim that the other can't be.
      2. Quote out of context
      3. Categorize their statements into some form of formula, thus trying to fight the argument by its appearance, not its content.

      To be fair, you got me on the -fviolate-ieee. I saw it on a /. comment and assumed it was correct. The full claim is that to get gcc to emit the same x86 FP code as the intel compiler, you have to use -ffast-math, which makes assumptions that can violate IEEE rules.

      The G5 is nice & quiet. A roomful of them at WWDC'03 was quiet enough for normal conversation at normal levels. With GCC's so-so optimizer, it beat out the Xeon, for which gcc's been more heavily optimized. We can see the real hard-core stuff with the IBM xlc vs ICC. Expect that when the G5 2Ghz start to ship in numbers.

      And when does describing scientific uses a cluster's make me superior??

      --
      Care about electronic freedom? Consider donating to the EFF!
    10. Re:I keep overestimating slashdot... by Anonymous Coward · · Score: 0

      gcc has been more heavily optimized for Xeon? You're kidding, right? No company is investing in making gcc optimize well for Intel processors. Apple has put developers into making it optimize for theirs.

      By the way, Xeon beat out G5 handily when Xeon was running icc compiled code and G5 was running gcc compiled code.

    11. Re:I keep overestimating slashdot... by bursch-X · · Score: 1

      >By the way, Xeon beat out G5 handily when Xeon was running icc compiled code and G5 was running gcc compiled code.

      We'll play that game again with xlc soon. Remember, the one who laughs last, laughs best.

      --
      There are two rules for success:
      1. Never tell everything you know.
    12. Re:I keep overestimating slashdot... by Anonymous Coward · · Score: 0

      Why does everybody keep offering a troll formula? I, myself, prefer the fighting uruk-hai formula...

  57. Mod parent down, blatant ignorance by Anonymous Coward · · Score: 2, Informative
    http://www.google.com/search?hl=en&ie=UTF-8&oe=UTF -8&q=opteron+systems&btnG=Google+Search

    Tons of multi-Opteron systems already being sold, and for some time now.

    Homeboy is mega-stupid.

  58. Re:Too expensive by Phroggy · · Score: 2, Insightful

    G5: Deliverable today

    Well, deliverable this weekend if you order 1,100 of them at a time; everybody else has to wait... but yeah.

    --
    $x='S24;r)>63/* h@<5+oZ)32"5cz';$me='phroggy'x$];
    $x=~y+ -xz+\0-Tx+;print$_^chop$me for split'',$x;
  59. TCO by CmdrTostado · · Score: 1

    They got a great deal from Apple on the machines, and when it falls of the bottom of the top 10 with the next release,they will have 1,100 machines to sell. Considering the excellent deal they got from Apple, plus a grant or 2 they picked up somewhere, they get great advertising for next to nothing. Of course I'm making this up....

  60. They are very quiet by Iowaguy · · Score: 2, Interesting

    Below is a link to show the noise of g5. (a movie). Apple did something called engineering (imagine that in a pc!), to put in many noise reducing features. So, the boxes may be bigger, but you get less power consumption and less noise. It is almost as if you pay extra money, and get extra features. Weird, I know....

    http://homepage.mac.com/aaronsteele/iMovieTheater2 0.html

    --
    "He who laughs last, didn't get the joke."-Cap
    1. Re:They are very quiet by jceaser · · Score: 1

      These people will never give apple a break. They want apple to give more but they are not willing to pay for it.

    2. Re:They are very quiet by negacao · · Score: 1

      what are you smoking?

      even a single processor G4 is.. LOUD.

      have you ever even opened one up? how can you say it's heat efficient?!

      FWIW, a dual G4 is about 6 db louder than my p4 2.66 w/ all the fixings. (keep in mind, +6db is 4x as loud.)

    3. Re:They are very quiet by Hawthorne01 · · Score: 1
      If we were talking about G4's, yes.

      But we're not.

      --
      "Only two things are infinite, the universe and human stupidity, and I'm not sure about the former."
    4. Re:They are very quiet by bursch-X · · Score: 1

      >how can you say it's heat efficient?!

      Open up a G4 case, what do you see? A passive heat sink, some fans, yep, that's the whole cooling.

      Now go to the next best 1 GHz P4, open the box. What do you see? Hey, they've put a vacuum cleaner in there and it sucks directly on the CPU! Active monster cooling without which the P4 would come to an immediate nuclear meltdown. That's actually why the codename of the P4 during developement was "Chernobyl"
      (OK, I made that one up, I admit ;-)

      Got it?

      --
      There are two rules for success:
      1. Never tell everything you know.
  61. The real news isn't the super computer by Chairboy · · Score: 2, Funny
    I think we can all agree that the shocking thing here wasn't that they were building the super computer as much as it was that a campus rumor was true!

    What will happen next, dogs and cats living together? Mass hysteria?

    1. Re:The real news isn't the super computer by SnappingTurtle · · Score: 1
      What will happen next, dogs and cats living together?

      Someone will throw a Pepsi machine out the window of Pritchard.

      --
      I've found that my posts don't format quite right w/o a sig.
  62. OS X 10.3 by Srsen · · Score: 4, Funny

    As it turns out, this is the minimum recommended system configuration to run OS X 10.3 Panther.

  63. Hahahah...right on right on. by FatSean · · Score: 0, Flamebait

    Fucking fanboys.

    --
    Blar.
  64. Beowulf Cluster by Luigi30 · · Score: 1

    Wow, imagine a Beowulf cluster of-- oh, too late.

    --
    503 Sig Unavailable

    The Signature could not be accessed. Please try again later or contact the administrator
  65. Not a personal hit by zerosignull · · Score: 1

    You know every time Apple Does something to improve it market share, company have a tendancy to want to do this ... its good buisiness sense, every PC troll comes out and puts em down. I never really thought PC users were generally self protentious peaople but for the most part they are. after reading this -> http://www.applelust.com/oped/amc/archives/amc0307 18.shtml article that goes into the facts, a word pc trolls dont like to hear, about the early dual g5-xeon comparison it made how bad you people really are clear to me. YES im sitting here running windows xp on my Athlon x86 PC but i , like unlike many i guess, have a mac sitting in the other room. At the end of the day the G5 is an emensly power full computer. even the dual's are used to genetics can u imagine what over one thousand of them can do? Just except the fact that appple has sold some computers and find more posts to troll

    1. Re:Not a personal hit by Anonymous Coward · · Score: 0

      I wish you had posted that from a mac. Safari has a built in spell-checker and if anyone needs that it's you. You might conceivably have written that with your nose, in which case I apologise.

    2. Re:Not a personal hit by zerosignull · · Score: 1

      You prove my point again by not attacking the FACTS of the matter but poor grammmer and spelling

  66. What's the long-term plan? by Anemomenous+Cowherd · · Score: 2, Insightful

    I have a sneaking suspicion that these computers aren't going to be used as a supercomputer for long. I bet they set this up, get on the supercomputer list, and then in six months or a year farm out the computers to use in computer labs around campus. Besides, I haven't heard a compelling reason why VT *needs* a supercomputer.

    1. Re:What's the long-term plan? by Anonymous Coward · · Score: 0

      Highly doubtful, less than 5% of our campus uses Apples. Take away the computers in the math emporium and I would be surprised to find 1100 macs on campus.

    2. Re:What's the long-term plan? by confused+one · · Score: 1
      VT has a long history in comp.sci. They're also a major engineering and sciences university. That's all the reason they need. They had an IBM manufactured supercomputer in '88 when I was there...

      You may be correct about putting the used machines into labs. However, I suspect they'll maintain their investment by replacing the nodes they remove with upgraded hardware.

    3. Re:What's the long-term plan? by RestiffBard · · Score: 1

      you are aware that VT is one of the leading agricultural research institutions right?

      Do you have any idea what it takes to accurately model a soybean field?

      --
      - /* dead coders leave no comments */
    4. Re:What's the long-term plan? by Anonymous Coward · · Score: 0

      If you don't know why they want it, you haven't been paying attention. They want to use this supercomputer to pull in research dollars. They believe that a few big grants could put the project in the black.

    5. Re:What's the long-term plan? by holzp · · Score: 1

      they need the computing power to figure out a better school mascot. theirs is so Hokie!

  67. No G5 For You by CPIMatt · · Score: 1

    This means that those of you who ordered your G5 early and expected to get one in early September, you are SOL. Your machine is goint VT. You have to wait another month before you get your machine.

    -Matt

  68. As an alumn... by Simkin1 · · Score: 2, Interesting

    As an alumn, I am irritated with the decision. As a cluster developer of 5 years, I'm highly irritated with the rational of the G5. It's one thing to develop a system for the intention of doing research, it's another to base a decision on "..delivered by Oct. 1..". The question you should be asking is, which is more important - getting on the list? or doing the research? Seems to me that there is a more cost effective solution, that provides higher capacity, greater throughput, and more overall compute capability at lower cost... I'd personally suggest VT slow down, rethink the cluster, and buy something that fits the needs of the school and research programs-therein.

    Side Note: While Tech has a great football team, the football program is (other than special discounts to students, and using the VT name) completely independent of the school. The football program is a business venture that does not interact with or require school permission, nor is it governed by the school boards that Steger answers to.

    Use Linux!

    1. Re:As an alumn... by Anonymous Coward · · Score: 0

      which is more important - getting on the list? or doing the research?

      Of course the research is more important, but the situation is much more complicated than you think. Since I'm involved in running a couple of big clusters in academia, here's how it works:

      1. You have $X and want a great computational resource.

      2. You want the best bang for the buck.

      3. You initiate discussions with several vendors, and discuss what discount the can give us, and what we can offer in return (place on top500, interviews with faculty praising the vendor, etc. "Fastest computer in state Y").

      4. Pick the alternative that gives us most for our money.

      It is really that simple. Sorry to rain on the parade, but VT almost certainly didn't first specify they wanted G5s. Apple knew they had a reasonably fast CPU, and wanted to get on the top500 list. The next list is published in november, and they next six months later when 2GHz G5 isn't as competitive. In other words: Apple needs the cluster to be benchmarked in october for the top500 list to get their publicity.

      Without that publicity they probably wouldn't have been able to get as good a deal from Apple, so the people in charge probably decided this was the best alternative.

      However, as both a Mac user and a computational scientist I wouldn't run OS X for completely different reasons: it is largely untested on this scale, and you want a cluster to be working 24/7/365 instead of fixing minor bugs in libraries, etc. Linux would be a better alternative simply because there are tens of thousands of cluster installations all over the world.

      Although, if I was negotiating for Apple I would demand they run OS X to get any significant discount (since it would give them good PR for the OS), so you never know... Microsoft sponsored a big cluster a couple of years ago on the condition that they run Windows. It was a complete disaster and the machine was essentially useless to scientists...

  69. Re:Virginia Tech by Anonymous Coward · · Score: 0

    Who would want to go to a school founded by an idiot like Jefferson anyway?

  70. Re:Virginia Tech by Anonymous Coward · · Score: 0

    UVA student?

    kekekek

  71. Re:Virginia Tech by Anonymous Coward · · Score: 0

    Please don't get us confused with the other supposed University in our state UVA.

  72. Re:plenty of games for mac by Jenolen · · Score: 0

    Who wants to pay for Photo$hop when you can get the GIMP running for free?

    --
    Karma is like sex. I can't remember the last time I had either of them.
  73. The G5 really is all that. by GPS+Pilot · · Score: 3, Interesting

    Read why here.

    --
    That that is is that that that that is not is not.
    1. Re:The G5 really is all that. by jjhlk · · Score: 1

      OT: Do you use linux on the mac? I just wanted to know how compatibility is for x86 linux apps on macs... Slick computers, if I can ever get the money.

  74. FP Performance??? by siskbc · · Score: 0, Troll
    Why don't you tell me how the hell I can use an Apple to get faster FP, smartguy. Does it optimize slashcode? Can it help me beat the subscribers? How does it get FP faster that the trollbot kiddies?

    All I'm saying is that if I lay out 3 large for an Apple and I *still* fail it, I'm going to be pissed. On the other hand, if I start beating out those GNAA turds for frost pist, it'd be worth it.

    --

    -Looking for a job as a materials chemist or multivariat

    1. Re:FP Performance??? by Anonymous Coward · · Score: 0

      well whoever modded you down totally missed the joke. *i* think you're funny. :)

  75. Re:you FUCKING PUDGE by MoP030 · · Score: 0, Flamebait

    i love reading with "Flamebait +6" modifier !!!

    --
    the most sexp i get is my paren-mode.
  76. Time for a new icon by 47Ronin · · Score: 2, Funny

    Now that we have G5s don't ya think it would be a better idea to use a G5 pic instead of the G4 icon? e.g. http://www.apple.com/g5processor/

    --
    Those who laugh at you for you having a Mac.. are the people who constantly call you to fix their PC.
  77. YHBT YHL HAND by Anonymous Coward · · Score: 0

    Apparently at least one moderator also has fallen for this one as well. People, please... Don't feed the trolls.

  78. Hmmmm by Anonymous Coward · · Score: 0

    Having actually built a G5 I hope the issue with the springy microphone/speaker sockets has been solved.

    1. Re:Hmmmm by commodoresloat · · Score: 1

      yeah, because we all know that Va Tech needs a supercomputer for students to do karaoke.

  79. imagine by Anonymous Coward · · Score: 0

    a beowulf cluster of dupes from retarded slashdot editors... oh wait.. that IS slashdot

  80. Re:you FUCKING PUDGE by Anonymous Coward · · Score: 0

    you, sir or madame, have just opened an unresealable can of WHOOP-ASS! prepare to be BLASTED!!!!!!!!!

  81. ... and add to foes list. by Anonymous Coward · · Score: 0

    Yup. He got the roles completely reversed.

    By the way, I'll have one of these :-o

    <drooool>

  82. Os license? by Frobozz0 · · Score: 1

    How the OS license on those go? I know with OS X you can get unlimited clients with each Xserve but I'm not sure how a cluster purchase works.

    If you seriously think about it, the OS cost is far beyond the hardware cost at time... especially on the Wintel side of things. It's an honest question, not a critcism. :-)

    --
    "Politicians find new names for institutions which under old names have become odious to the people."
  83. Re:Virginia Tech by splatter · · Score: 1

    Wrong again... sherlock Thats UVA.. as to VA being a police state or VTech being a red neck campus I wouldn't know I left VA for college, but I don't think one of the top rated science schools in states is all that bad.

    DP

    --
    "(I) have this unfortunate condition that causes me not to believe a single thing any politician says when a mic's on.
  84. cool, but... by V_drive · · Score: 1

    what about photoshop benchmarks?

    ::ducks::

    --
    char *mySig;
  85. Umm... OK.... by greymond · · Score: 2, Informative

    So I read this "The project comes at a time when the university's academic departments are struggling to fulfill students' educational needs in the wake of a $72 million reduction in state support."

    and then I wonder why you would spend $5mil dollars over the next 5 years to build a supercomputer? It seems like a better idea would be to reach out to the slahsdot/linux communities and see what kind of equipment they could get donated/free and then build a semi-super computer with that - or hell even just buy a shitload of cheap pc's to do it with....

    maybe i'm just missing something...

    1. Re:Umm... OK.... by hayne · · Score: 2, Insightful
      and then I wonder why you would spend $5mil dollars over the next 5 years to build a supercomputer? It seems like a better idea would be to reach out to the slahsdot/linux communities and see what kind of equipment they could get donated/free and then build a semi-super computer with that - or hell even just buy a shitload of cheap pc's to do it with.... maybe i'm just missing something...

      Yes - you are missing that they want to have a top-notch supercomputer and they want it now, and they don't want to be fiddling with a mismatched bunch of cheap or donated equipment, and that the funding for this likely comes out of a different purse than that used for regular ongoing expenses.

    2. Re:Umm... OK.... by Anonymous Coward · · Score: 0

      There's also the issue that research is separate from undergrad education. Different source of funding, different focus, and living in a parallel university where the undergrads and the departments that cater to them are invisible.

      Real research requires real resources, and mismatched donations from Slashdotter closests is not going to cut it.

    3. Re:Umm... OK.... by Anonymous Coward · · Score: 0

      You are missing the point that they think they can make back the entire cost of the project with the research grants that it will theoretically pull in. So the money isn't being drained out of anybody's pockets, in the end.

    4. Re:Umm... OK.... by Anonymous Coward · · Score: 0

      The money isn't being drained out of anybody's pockets... unless you count taxpayers' pockets. We're most likely talking about federal research grants, after all.

  86. No increase, decrease! by TekkaDon · · Score: 1

    Why don't you zip it and please read the facts from the horse's mouth:

    ["Virginia Tech will have one of the top ranked supercomputing facilities in the world, supporting significant "big science" research. It is anticipated that Virginia Tech will realize at least a five to one return on this investment in terms of annual research grant and contract activity," says Glenda Scales, assistant dean of computing and distance learning at Virginia Tech.]

    Let me say that again: Five to one return of investment projected. At least. If they are saying this outloud, I can assure you Ms. Scales knows what the hell she's talking about.

    You can find it here

  87. FIRST reliable supercomputing facility... by TekkaDon · · Score: 5, Informative

    Or so they claim here. It seems they have all their bases covered and don't give a damn about ECC for a reason.

    [Srinidhi Varadarajan, an assistant professor of computer science at Virginia Tech, and Jason Lockhart, director of the College of Engineering's High Performance Computing and Technology Innovation, initiated the venture at Virginia Tech. Varadarajan is an expert in reliability, a key issue in successfully exploiting terascale computing.]

    They keep on going:

    [Component failures are endemic to any large-scale computational resource. While previous generations of supercomputers engineered reliability into systems hardware, today's high performance computing environments are based on inexpensive clusters of commodity components, with no systemic solution for the reliability of total machine.]

    And now for the solution for your reliability problem.

    [Virginia Tech has the first comprehensive solution to the problem of transparent fault tolerance, which enables large-scale supercomputers to mask hardware, operating system and software failures - a decades old problem. It's a software program called Deja vu, designed by Varadarajan. He also integrated the software with Apple's G5s. This work will enable the terascale computing facility to operate as the first reliable supercomputing facility, according to Varadarajan, a National Science Foundation Faculty Early Career Development Program (CAREER) Award recipient.]

    So maybe, just maybe, you and other people could:

    1. READ before posting.

    2. Then READ a little more.

    3. Did I say READ already.

    -sigh- Whatever.
    1. Re:FIRST reliable supercomputing facility... by FlowerPotAdmin · · Score: 0, Troll

      Deja vu

      Sounds like checkpointing to me. And in any case, while the management program can detect things like fail-stop faults of system components, the only sure way to guard against a few random flipped bits (without ECC) is to duplicate the entire computation.

      I don't give a crap what VT's news department claims. READing isn't going to be of much help if you neglect to THINK.

      --
      -Justin
      That's enough posting for now lads, there're trolls afoot.
    2. Re:FIRST reliable supercomputing facility... by bored · · Score: 1

      The entire computation may != all memory. They may be able to CRC chunks of inactive/readonly memory and then duplicate only the pieces under active computation. Seems like a lot of effort to save a couple $ by buying non ECC memory to me though...

      Maybe the guy has some breakthrough scheme. More than likely though I would bet it has some significant flaw. They are probably using Mac's not because they are the best solution but rather because they will be able to use them in the public computer labs in a year..

    3. Re:FIRST reliable supercomputing facility... by Anonymous Coward · · Score: 0

      Since when "THINK" means:

      "Sounds like..."

      Thought that was more like "SPECULATING"

    4. Re:FIRST reliable supercomputing facility... by Anonymous Coward · · Score: 0

      The entire computation may != all memory. They may be able to CRC chunks of inactive/readonly memory and then duplicate only the pieces under active computation. Seems like a lot of effort to save a couple $ by buying non ECC memory to me though...

      The problem isn't that it is difficult - CRC is trivial in practice, but it would ruin your performance. The G5 can send a couple of *gigabytes* per second to/from memory, and you would need to recalculate the checksums every time you write the new results.

      Most scientific applications are limited both by floating-point operations and memory throughput. How does this affect them?

      Alternative 1: Copy all them memory. This will cut the memory bandwidth in half, not to mention that the CPU will be idle while we wait for the copy to complete. In other words, not an option.

      Alternative 2: Calculate checksums. Even if we use Altivec this would take all the CPU power available just for checksums, and then you don't have the altivec unit anymore. (Not to mention that the CPU will still be idle while calculating checksums.)

      Checkpointing a job isn't an alternative to ECC. It helps if your node crashes, since you can restart from an earlier timestep. It will NOT detect random memory errors. Those will just silently corrupt your data.

      The G5 is great, bla, bla, bla. I want one, and we're replacing all our laptops in the lab with Powerbooks. But it isn't a coincidence eveybody uses ECC in big servers or clusters.

    5. Re:FIRST reliable supercomputing facility... by fitten · · Score: 1

      Not to mention that the checksum calculations will destroy the cache and your performance will go into the crapper. I can't imagine this as being a viable option.

      The only real option is to have a voter system where multiple nodes (or subclusters) all calculate the same thing at the same time. You compare all the answers and take the answer set that has the most nodes (or subclusters) with that result.

    6. Re:FIRST reliable supercomputing facility... by Anonymous Coward · · Score: 0

      Hi there, I'm the AC who first posted wondering about building a cluster of machines that don't have ECC (or even parity checking). (No, really!)

      I hate to break it to you, but the Deja vu software has nothing at all to do with ECC. It's an implementation of MPI. In other words, it's a library that lets you fairly simply write C/C++/Fortran programs which run on multiple computers simultaneously, communicating with each other via TCP/IP, or a number of other "specialist" interconnects. There's nothing special about MPI: lam (http://www.lam-mpi.org) is a great, free implementation.

      How is Deja vu different?

      Suppose you know a computer is about to blow up (perhaps a fan has failed and the CPU temperatures are starting to climb) - Deja vu will let you silently fail-over the machine to a "hot spare". Other MPI implementations don't let you do this.

      I'm sure you can understand that this has nothing to do with preventing (or even detecting) data corruption. And how could it? Deja vu, after all, is a piece of software running on the machines. What if one of _its_ bits is corrupted? Obviously, it's not going to do whatever is required to keep the machine running correctly, but crash in a heap. Now the odds of this happening are low, but vastly greater on an Apple G5 system than on most any other kind of computer that people build clusters with.

      Anyway, while reading is great (thanks for pointing me to Deja vu), thinking for yourself is even better. Give it a try, I'm sure you'll like it.

    7. Re:FIRST reliable supercomputing facility... by Anonymous Coward · · Score: 0

      Maybe the guy has some breakthrough scheme. More than likely though I would bet it has some significant flaw.

      Yes, I'm sure you are much smarter then the guy they hired to build this facility. I keep forgetting that slashdot users know everything!! Why didn't they just ask you to build it for them in the first place? How stupid of them!

    8. Re:FIRST reliable supercomputing facility... by Khlatu_Barada_Nicto · · Score: 0
      The only real option is to have a voter system where multiple nodes (or subclusters) all calculate the same thing at the same time. You compare all the answers and take the answer set that has the most nodes (or subclusters) with that result.

      But would such a system perform acceptably in Florida?

  88. Damn, by Gannoc · · Score: 1


    The moderators are getting so bad, they're making the submitters produce dupes for them in advance...

  89. Re:Virginia Tech by Anonymous Coward · · Score: 0

    Q: How many Virginia Tech students does it take to change a light bulb?

    A: Three - one to change the bulb, and two to discuss how they did it as well as a UVA student.

  90. Re:They want to play duck hunt, idiots! by Anonymous Coward · · Score: 0

    -1 troll?!

    C'mon, that's over 40 hours of keep your pants dry entertainment!

  91. Sigh... by mbessey · · Score: 1

    Look, it's really simple. The G4 is loud simply because they didn't put in the engineering effort to make it quiet. The G5 on the other hand, is really really quiet. I was just using one today, so I know whereof I speak.

    -Mark

  92. I wonder how much *wind* this will create by slart42 · · Score: 1

    considering the G5's 'smart' airflow design, engineering the case like one big air channel, I would imagine that 1100 of these, if aligned correctly, could sum up to quite a nice breeze...

  93. Is it research? by j.leidner · · Score: 0, Troll
    Is it considered research now to buy a desktop computer from a shop and abuse it as a server?

    Why make so much fuss about buying a bunch of Macs off the shelf? My grandmother could do that if she wanted. The fact that they want to be finished "building" it (= connecting the cables) seems to be simply that they want to make it into the Top-500 list with $$$ rather than new research.

    What they will find out is that the G5 was never built to be run as a server, despite its server-like performance. And 1100 Mac end users (for whose desktop the machine WAS built) will perhaps have to wait longer until they can buy theirs.

    Back in the old days any serious computer engineer would refuse to buy off-the-shelf machines and rather build their own -- I believe research should lead the crowd wrt innovation, not follow it! At least they should hack the Linux PPC kernel or build a new cool software layer to contribute something rather than just consume.

    Just my two cents worth...

  94. Ray Rah VPI! by SamBaughman · · Score: 1
    Maybe VT can use the new computers to beef up their web site.
    May I mod that completely unneccessary stab at my alma mater as -1 Flamebait?

    While I was a student at Virginia Tech ('94-'99, including a year of co-op) as a Computer Engineering student, I was always impressed by the diversity of computing resources throughout the university. Each department generally sided with a single environment (DECstations to FreeBSD PCs for CS, Windows (AutoCAD) for Engineering, Macintosh for Math & English), although all were supported. Even all the way back to the mainframe and its dumb terminals, which students used to sign up for classes. The general student at Virginia Tech learned, by necessity, how to do the basics across a variety of systems.

    I'm happy to see Virginia Tech continue its push forward. "Commodity supercomputers" through clustering almost always refer to Intel-architecture systems. Why not Apples as well? It's a brilliant move forward, not only for the computational power this involves but also from a P.R. perspective... would all those high-school techies have heard about this if they chose an Intel architecture solution? And Apple will get good P.R. as the building blocks behind a supercomputer.
  95. IB vs. Myrinet, Quadrics by Troy+Baer · · Score: 1
    I can't seem to find the quote from any of the articles right now, but VT is planning on using an Infiniband interconnect from Mellanox. While I don't know the relative price points, they are touting the fact that this is a high-speed interconnect that's faster than Myrinet or Quadrics at a fraction of the cost. I can't say for sure, since the Infiniband cluster we're helping to build at Stanford is not yet assembled.

    The numbers I've seen seem to suggest that a 128-host IB fabric is only slightly more expensive than a Myrinet fabric of the same size, with similar latency characteristics for MPI programs and about 3x the bandwidth per host. Quadrics has better latency characteristics than IB, but about half the per-host bandwidth; I'm not sure about relative cost.

    What I'll be very curious to see is how Apple and VT scales an IB fabric out to 1100 hosts. They'll either have to buy a boatload of "backplane" switches or sacrifice bisection bandwidth across the system as a whole. That'll probably depend on whether they have big bisection-bandwidth-bound codes, like parallel FFTs...

    --Troy
    --
    "My life's work has been to prompt others... and be forgotten." --Cyrano de Bergerac
  96. Ob. Latin Grammar Nazi post by Anonymous Coward · · Score: 0
    that'd be 'alumnus'.

    If you're going to be one, you may as well spell it properly.

  97. No If's And's or But's: The PPC970 is Super Cool! by reporter · · Score: 0, Troll
    The Apple G5 is powered by the PowerPC 970. It has one unusual feature that is not shared by the Athlon64 (AMD), the Pentium 4 (Intel), and the Itanium 2 (Intel). Most of the engineers, technicians, and support people who worked on the PowerPC 970 were not H-1B employees. IBM has a strict policy of not employing H-1B employees except in positions that require a Ph. D.

    The PowerPC 970 shatters the myth that you absolutely need H-1Bs because supposedly there is not enough talented American engineers to do the work. By nearly all metrics, the PowerPC 970 is competitive with the very best processors produced by H-1B-dominated companies like AMD and Athlon.

    The Apple G5 may be slightly more expensive than an equivalent system based on an AMD/Intel microprocessor, but the Apple G5 is worth cost. Buying the Apple G5 supports the more traditional American work environment at IBM as opposed to the brutal, cutthroat environment at Intel.

    ... from the desk of the reporter

  98. What will it be used for? by MacGod · · Score: 0, Troll
    As for what this Super-duper-computer will be used for:

    As The Apple Turns reports here, that the supercomputer cluster's main purpose will be to "impress the living crap out of everyone."

    Just thought you'd all like to know

    --
    "Reality is merely an illusion, albeit a very persistent one " -Albert Einstein
    1. Re:What will it be used for? by reiggin · · Score: 1
      Why has this modded "troll"? He's merely linking to a humorous quip from a semi-satirical Mac site. It's okay if you don't think it's funny but the very fact that this guy still reads AtAT proves that he is not an anti-apple troll or any other kind of troll for that matter.

      Zealot, perhaps. Troll, never.

      Hmmm. "Zealot." Me thinks we need a new mod category...... If only because it'd REALLY help my karma.

      Going back to lurking as a very proud mac zealot......

    2. Re:What will it be used for? by Anonymous Coward · · Score: 0

      Doesn't matter file under "JAC", Just Another Cluster. I am amazed at the hype this stuff draws
      and the funding folks manage to suck up for it.

      This is supercomputing with a little "s".

  99. MOD PARENT UP! by sockit2me9000 · · Score: 1

    text

  100. OOPS! by frank_adrian314159 · · Score: 1
    Nothing like a cut and paste error to spoil a joke!

    The setup should have been:

    the fan configuration will make it extrordinarily loud

    and the punch line (which I did get right) is:

    On the plus side, the aeronautics department can use it simultaneously for a wind tunnel!

    But who knows! Maybe the incorrect one works, too!

    --
    That is all.
  101. Apples revenue: by Anonymous Coward · · Score: 0

    Apples total sales for the year to date: 1100

    1. Re:Apples revenue: by knghtrider · · Score: 1

      No, 1102--you forgot the two boxes the geeks at VA Tech bought to test the environment. And lest you think Apple provided them for free; how do you think the group proved they spent the grant money? Think about the professor in 'Real Genius' for a minute.. LOL..

      --
      In America today you can murder land for private profit. You can leave the corpse for all to see, and nobody calls the c
    2. Re:Apples revenue: by Hoser+McMoose · · Score: 1


      <heard from the bowels of Intel's headquarters>

      Damn, Apple beat our Itanium shipments by 100 machines!

  102. Apple marketroid by penguin7of9 · · Score: 0, Troll

    Wonderful: three pointers to Apple's web site, pointing to pages with slick corporate "interviews". Do you actually work for Apple or are you just insanely zealous?

    1. Re:Apple marketroid by Anonymous Coward · · Score: 4, Informative

      Well I don't know about other science work but my dad does AIDS research and they use Macs exclusively in their labs for all of their model simulations and experiments. There's apparently a lot of biology research software that is available only on the Apple platform as well. I guess maybe because of their ties to educational institutions over the years?

    2. Re:Apple marketroid by BWJones · · Score: 3, Interesting

      Wonderful: three pointers to Apple's web site, pointing to pages with slick corporate "interviews". Do you actually work for Apple or are you just insanely zealous?

      There are an awful lot of scientists using Macs for their research and work. I use them almost exclusively now after retiring my SGI's in favor of the OS X boxes and judging from the meetings I attend, I would say Macs have anywhere from 10-40% penetrance in science depending upon the subfield. For instance the last vision meeting I attended (ARVO, the big one for the vision research community), there were Powerbooks and iBooks everywhere. Probably a good 33% of the laptops I saw.

      --
      Visit Jonesblog and say hello.
    3. Re:Apple marketroid by penguin7of9 · · Score: 1

      Yes, and there is a lot of research software that's only available for Windows, and there is a lot of research software that's available only for UNIX. You see, sometimes people write software for specific platforms.

  103. ECC - why is this important? by zpok · · Score: 1

    OK, apart from being a nice thing to flame people with, what is this ECC talk?

    Why is it so important?

    Are there other "supercomputers" running without it?

    Is there a good reason to disregard it and go with G5 anyway?

    --
    I think, therefore I am...I think.
  104. Specs from an involved student... by Coocha · · Score: 5, Informative

    My boss here at VT is a volunteer for this project... they've been designing and building rackmount shelf-type units to store all these new G5s, as well as helping with the cooling system. Here's some info he gave me.

    The cluster will eventually run Mac OS 10.27... he said eventually, and Jason Lockhart, the project leader, is a friend and fellow Linux geek of mine (please don't hammer his inbox ;-), so there's a chance that he might use some PPC distro at some point.

    Interconnectivity will be done with Cisco equipment, among the onboard gigabit LANs. Infiniband cards will also eventually be installed for 10 Gbit throughput.

    You guys can offer alternative solutions and troll this as much as you want, but this is what VT is going with. In my opinion, it's not a bad choice... the New IBM PPC chipset is balls-to-the-wall computing, and Apple's 'stock' offerings in the G5 (Gbit ethernet, serial ATA, etc.) are all strong selling points. The fact that this cluster is intended for intense vector and matrix-based algorithms is another bonus, b/c of the PPC vector processing unit.

    Apparently Apple shifted us up to the top of their production ladder, in order to make the contract, thereby extending the wait times for consumers itching for a G5... I find that a little humorous. Can't wait to see gigaflop statistics!!

    --
    May the threads progress competently.
    1. Re:Specs from an involved student... by Wesley+Felter · · Score: 1

      Let's hope you can get more than 600Mbps on the G5's Ethernet port.

    2. Re:Specs from an involved student... by bspath1 · · Score: 1

      What does the onboard Ethernet have to do with the Infiniband cards that they're planning on installing?

    3. Re:Specs from an involved student... by Wesley+Felter · · Score: 1

      From Coocha's post:

      Interconnectivity will be done with Cisco equipment, among the onboard gigabit LANs. Infiniband cards will also eventually be installed for 10 Gbit throughput. (emphasis added)

      I'm just pointing out that since they'll be relying on the onboard Ethernet for now, hopefully the G5 has better Ethernet performance than previous Macs. You're right that it won't matter once they get InfiniBand working.

    4. Re:Specs from an involved student... by Coocha · · Score: 1

      In our [multimedia] lab, we use the G4 silver's onboard gigabit to connect to a 2-terabyte raid for uncompressed video storage. It sure ain't 'true-blue' gigabit throughput, but it peaks at about 200Mbyte/sec. I'm crossing my fingers that Jason, with his *nix experience, will be able to do the necessary tweaking to the systems and whatnot to boost that beyond stock OS X capabilities.

      No one has told me why the timeline is so short for getting this thing up and running... probably has something to do with the grant money, and a related project needing flops. But hopefully their conversion to the infiniband gear will be sooner, rather than later.

      Thanks to moderators for giving my parent my 1st +5... I divulged all I knew, and I'm glad it satisfied the masses!

      --
      May the threads progress competently.
    5. Re:Specs from an involved student... by SeanAhern · · Score: 1

      No one has told me why the timeline is so short for getting this thing up and running...

      I hear they are rushing to try to get included in this year's Top 500 Supercomputers list.

  105. Re:ECC - no longer an issue!!! by zpok · · Score: 1

    This here poster has answered the question already.

    I'd say, mod him up, flog me for not reading this myself and go flame some ECC fanboys... (did I just say that?)

    Seriously though, I thought ECC - or something alike - had to be important, reading about it, but OTOH couldn't imagine this university (with quite a track record already) not coming up with this themselves.

    It sort of confirms my views on *some* wannabe-scientists on this list.

    Having a degree doesn't equal having common sense. I like playing high and mighty like any other guy, but come on... This is largely a repeat article, it should have given some people the time to set their facts straight.

    --
    I think, therefore I am...I think.
  106. Re:PC Upgrade Woes by zpok · · Score: 1

    Woohoohooo

    I always found this post to be absolutely idiotic. It still is of course, but almost pissed my pants laughing :-))))

    --
    I think, therefore I am...I think.
  107. Rumors getting closer by Kalak · · Score: 3, Interesting

    In talking to the person who is recruiting me to help lug the computers around when they arrive, the OS is to be OS X 10.2.7 on arrival, with plans to upgrade to Panther upon it's release. Straight out-of the box releases, with NetBoot planned to be used to distribute the images to each computer. This contradicts the rumors I've heard before, but is closer to a source who is on the planning team, who is too damn busy to talk to a luser like me.

    Those who are possible volunteer recruits, there is an info session in Andrews ISB in the Corp. Research Center at 7:30 tonight and tomonrrow night (same presentation both nights). You *cannot* be on wage for VT to be elegible. I'm not sure if GAs count as this, since I'm not one, I didn't check.

    --
    I am, and always will be, an idiot. Karma: Coma (mostly effected by .hack)
  108. MOD PARENT UP! by Anonymous Coward · · Score: 0

    HERE! HERE! This guy is right on, like his remark on EEC up the thread. It's all in the PR note, damnit.

    I guess we all should READ the available information before posting around like headless chickens.

  109. Tech Tech VPI! by tube013 · · Score: 1

    Go Hokies

    I was down there last week for the game, and the school is definitely transforming. also the links in the article point to off VT network sites, so is the comment warrented?

  110. The answer to your problem by WankersRevenge · · Score: 1

    If you want, I can let you borrow my ex-girlfriend. Just put her in the corner of the room and she'll solve that cooling problem of yours in no time.

  111. "Magic Bus" (See Apple page) by vanza · · Score: 1

    Is that where the magic smoke runs through so the computer can work?

    --
    Marcelo Vanzin
  112. Hmmm by Anonymous Coward · · Score: 0

    Their is one problem with the new 65's.
    THEY ARE UGLY!

    1. Re:Hmmm by reiggin · · Score: 1
      Their is one problem with the new 65's.
      Okay. And there is at least one problem with your perception. I'd say that if that logo on the side of the machine looks like a "65" to you, then your opinion that the entire machine is "ugly" is easily disputable.

      I suggest LASEK/LASIK.

  113. $26 million? by Agent+R · · Score: 1

    Berkeley did it on $500 grand with their SETI@home project.

    --
    !@#$% whole-grain cereal. When I want fiber, I eat some wicker furniture. - G. Carlin
  114. Ummm... by Anonymous Coward · · Score: 0

    ...while i agree with you, it doesn't pan well for your argument that you sited three pages from Apple.

    Remember what site your posting on? That might have worked on n00b.com, but this is /. and we, dare I say, are a little more cunning that that.

  115. Good question Sir... by Anonymous Coward · · Score: 0

    But I want to know...

    ...will we have to wait for Mac-Doom III on this G5 system to get decent frame rates?

  116. dual use for these machines? by esome · · Score: 2, Interesting

    would it be possible to use the machines both as lab machines for students AND as a cluster? I mean does the gear that networks them necessarily prevent them form being used as individual machines? sorry for the n00b question but it sure seems like VT would get a lot more for their money that way.

    1. Re:dual use for these machines? by Gilmoure · · Score: 1

      There's some software called Pooch that lets you run parallel clusters across PPC Macs. It's pretty minimal but you can still use the Macs for regular work and then run parallel software on them when they're not being used.

      --
      I drank what? -- Socrates
  117. Old Sun 4c hardware by maynard · · Score: 1

    Dude, look at an old sparc sometime. Sparc 1/1+/2 had 16 ram slots, circa 1990. Of course, you had to fill 4 at a time. The max is 128 MB i think.

    Not unless you could find a 30 pin SIMM in 32MB/four density, which I've never seen; the biggest being 16MB/four. I still have several Sun 3/80's, Sparc 1s, 2s, 5s, and 20s sitting in my basement. And the 20s I still consider a perfectly usable workstation. They were great computers in their time and I still have many fond memories hacking those Suns. But *sigh* it's time to clean out my basement. :) --M

  118. As a current grad student... by J0ey4 · · Score: 1

    I would advise you to have a little more faith in your alma mater....whoever told you there is a:

    "more cost effective solution, that provides higher capacity, greater throughput, and more overall compute capability at lower cost..."

    was smoking crack. If it was yourself than I respectfully disagree (and would like to see said solution).

    Each PPC970 has two independent FPU, each of which is capable of executing a fused multiply/add instruction. This allows the PPC970 to function as if it had FOUR FPU's when doing all that nifty matrix math that makes up a large part of scientific calculations. This puts the machine's peak performance at over 17 Teraflops. If you don't believe me read the article over at arstechnica, or just look at the PPC970 documentation. Not to mention that the Altivec SIMD unit (for any repetitive single precision calculations) SMOKES AMD & Intel's vector solutions. Also the Mellanox Infiniband communication fabric has ridiculously low latencies and extremely high bandwidth..beats Myrinet on both counts from what I've seen...

    The long and the short of it is that a dual-Opteron built cluster would have to have significantly more nodes (== more expensive) to deliver the same peak (or average for that matter) performance, and would take longer to be built.

    Apple wins on price.
    Apple wins on performance.
    Apple wins on delivery time.

    I'm sure the decision to use the G5 was not made lightly or frivoulusly, and that all the options were carefully weighed before choosing the one made the most sense.

  119. What da hell will make /. change its G4 thumbnail? by Anonymous Coward · · Score: 1, Funny

    Come on /. Would you please grab a damn G5 thingie offa something somewheres and quit using G4 thumbnails in Apple stories? It sticks out like a green sore thumb and it's embarrassing for /. Time to roll with the changes.

  120. Re:ECC - no longer an issue!!! by Anonymous Coward · · Score: 0

    Seriously though, I thought ECC - or something alike - had to be important, reading about it, but OTOH couldn't imagine this university (with quite a track record already) not coming up with this themselves.

    Sigh... let's separate two things.

    1. Checkpointing. This means saving the state of your job in regular intervals. It can be done by the program, or automatically by the operating system. The "Deja vu" approach mentioned is checkpointing. It isn't new - SGI has supported it for a decade, and CRAY another decade before that. Checkpointing helps if the CPU crashes or if there is a power outage.

    2. Memory errors. These are random, mainly caused by alpha-particles from space. ECC detects and correct this. Silent random bit errors is horrible, since they won't crash your machine - you won't even know the error happened without ECC.

    Sorry, it doesn't matter how smart the "Deja vu" program is - It cannot detect an undetectable error. The only way to be sure is to run all your simulations twice and make sure the results match...

  121. Re:Too expensive by Anonymous Coward · · Score: 0

    Erm, your not paying for it ARE YOU

  122. Re:But WHY? by Gogo+Dodo · · Score: 1

    The PowerPC 970 is based on the Power4. You need to check your facts. (Hint: visit the IBM website)

  123. Re:ECC - no longer an issue!!! by zpok · · Score: 1

    Thanks.

    Some questions, if you feel like answering, that is.

    Any reason why Apple doesn't support this ECC memory apart from being a bit over the top for simple desktops?

    Can they be custom installed (sort of like "look here, I'm going to build a supercomputer instead of doing this photoshop stuff, so put those ECC thingies in there, ok?") or is it plain not supported, not possible?

    Is ECC memory readily supported by other manufacturers (apart from SGI and CRAY)?

    And the obvious one, why do you think a place like Virginia U. overlooks this while they - and the people responsible - obviously have firsthand experience in this?

    --
    I think, therefore I am...I think.
  124. Re:ECC - no longer an issue!!! by Anonymous Coward · · Score: 0

    Why didn't they mind missing ECC? My guesses:

    1) It's a fuck-up. Perhaps the grant money really really did have to be spent right away and Apple was the only vendor able to deliver on short notice. Maybe the researchers figure they are going to run the simulations twice, after all an effectively 550-processor machine is a lot better than getting nothing if the grant couldn't be cashed for whatever reason.

    2) Apple has undocumented ECC support on their chipsets which they figure they can get up and running in time. This case is most likely if the "no ECC support" line is just a marketing decision to differentiate G5 desktops from future G5 servers... It would be insane if they're now trying to make first-batch G5 machines use a feature that hasn't been previously tested or worse, a feature that failed testing and was not documented because of that.

    3) The "reliable clustering" software they have does something extremely clever to work around ECC errors. Say, they could run a background memory-checksum process and in case of checksum failure backtrack to the latest checkpoint on the failing machine and on all machines that have used results provided by that machine. It seems unlikely that something like this would be sufficiently efficient and worth the extra complexity, but perhaps they have worked out the details and it does work out.

    4) The simulations they run aren't fussy about some random perturbations here and there. Perhaps they even actively add random perturbations for some physical reason. Then, if 99.999% of memory is used for numerical simulation data, the probability of a memory error hitting the OS, application code, or critical pointer data is small enough.

  125. I use OS X for science by Anonymous Coward · · Score: 0

    BlastX is sweet with a G4

  126. It's all well and good, but... by jonathan_the_ninja · · Score: 1

    Why does the G5 say "G4" on it?

    --
    I love NetHack.
  127. Wow...what a shitty Troll... by FatSean · · Score: 1

    Better luck next time.

    --
    Blar.
  128. Re:What the hell do you do with a head this thick by dave1212 · · Score: 1

    ...the right number of buttons.. the right number of buttons for a beginning user is one. (1)

    I, like many other Mac users, purchased a 3 button mouse almost immediately after getting each Mac I've owned.

    Do you think it's time to get your head of that hole and stop with the "number of mouse buttons" complaints?

    I don't understand why you just want to chime in and miss the joke entirely..

    good luck with that, eh? maybe you'll eventually save up enough to get a Mac and not be so bitter. ;)

  129. Re:Maybe an Insiders perspective on VT's Decision. by Mr.+Darl+McBride · · Score: 1

    Moderated at:
    -1 Doesn't Reaffirm My World View

  130. But not everyone can run Deja Vu! by slackbp · · Score: 1

    As far as I can tell, the press release you referenced wasn't mentioned in the original Slashdot postings. It's not fair to criticize someone for not reading something they didn't know existed, is it? In any case, while I'm sure the Deja Vu software is useful, it doesn't help those of us who aren't setting up clusters--there's still a need for reliable hardware. Check out http://cr.yp.to/hardware/ecc.html for reasons why.