Catch up on stories from the past week (and beyond) at the Slashdot story archive

 



Forgot your password?
typodupeerror
×
The Internet Hardware Technology

Weta Digital Supercomputer For Hire 184

sushi writes "NZ's Stuff news site is reporting: 'Peter Jackson's special effects shop Weta Digital has teamed ... to establish a world-class supercomputing facility in Wellington which will be rented out to clients worldwide.' Currently comprising 504 IBM blade servers, each of which contains two 2.8 Gigahertz Intel Xeon processors, 6 Gigabytes of memory and 40 Gigabytes of storage, and ranked 80th in the top 500 supercomputers, they are intending to upgrade into the top 10. Also covered at the Australian Financial Review."
This discussion has been archived. No new comments can be posted.

Weta Digital Supercomputer For Hire

Comments Filter:
  • So... (Score:3, Funny)

    by iamdrscience ( 541136 ) on Tuesday September 07, 2004 @02:54AM (#10174820) Homepage
    Are we going to see a "Meet the Feebles part 2" or what?
  • Dupe? (Score:5, Informative)

    by Anonymous Coward on Tuesday September 07, 2004 @02:55AM (#10174824)
  • Awesome (Score:3, Funny)

    by iamdrscience ( 541136 ) on Tuesday September 07, 2004 @02:58AM (#10174830) Homepage
    So now instead of curing diseases, finding large mersenne primes or discovering inteligent life forms we can get hyper realisitc renderings of Gollum's eyelash crust. Spectacular.
    • Re:Awesome (Score:1, Funny)

      by Mindcry ( 596198 )
      its not like you really wanted to cure diseases anyways ;) now bow down before the greatest generation of eyelash crust the world has ever seen.
    • There was a song by the Gang of Four in the "ye olde days" which went. The problem of leisure, what to do for pleasure. In this day and age entertainment is more important then finding primes, dicovering intelligent life forms, or curing disease for 90+ percent of the western world. BTW does anybody know if "entertainment" is still in print?
      • Entertainment! was re-released by EMI circa 95 or so with bonus tracks from an early LP slapped on the end and should be available through better CD stores. Glad to hear someone else out there still remembers GoF - they managed to make it into one of my recent password generation schemes.

        I highly recommend checking out GoF member Dave Allen's Elastic Purejoy self titled album (from 1994). It ranges from very GoF inspired dance-rock to grunge rock to psychedelia and everywhere inbetween. Between songs wi
        • Sometimes I worry that the only music I buy is from 20 fucking years ago or older. I want to be surprised, shocked, and awestruck like the first time I heard "Entertainment" or "bitches brew" or "virgin beauty". What happened? Are people like Miles davis, Ornette coleman and GOF still around? Are we all doomed to live in United States of Mediocracy.

          I have been really reluctant to buy old CDs I used to have to try and relive old the days but I will have to make another exception for GOF I guess.
    • Re:Awesome (Score:5, Insightful)

      by Jerry Talton ( 220872 ) on Tuesday September 07, 2004 @03:40AM (#10175016) Homepage
      It really cracks me up that you rate finding large mersenne primes in the same category as curing disease or discovering other intelligent life forms. Talk about something with no practical applications...
    • Re:Awesome (Score:2, Insightful)

      or discovering inteligent life

      Or, you know, a spell-cheeker...

    • Sounds good to me.

      Better than:

      'We have foundz ze nth dimenzion of ze nd dimenzion of the nd dimenzion of ze univerz'

      How will this bring joy to our lives and prevent world poverty?

      'We have foundz ze nth dimenzion......'

    • Re:Awesome (Score:3, Insightful)

      by LoudMusic ( 199347 )
      You mention using it for the standard distributed projects, but what about the computers that are already running these projects? Are they not part of a super computer? A type of cluster? Surely the fellas (and ladies?) at distributed.net and / or the SETI@Home crew could write up a simple distributed RMAX app to test how much CPU time is available on the internet. Submit those numbers to the top500.org and see where you end up. I bet it'd put the Earth Simulator to shame.

      Now, for 'real time', it'd be shat
  • top 500 ! (Score:5, Informative)

    by phreakv6 ( 760152 ) <phreakv6@gma i l . com> on Tuesday September 07, 2004 @02:59AM (#10174832) Homepage
    This is the top 500 [top500.org]
  • Correction? (Score:5, Insightful)

    by iamdrscience ( 541136 ) on Tuesday September 07, 2004 @03:03AM (#10174854) Homepage
    Hmm... if you look at the top 500 [top500.org] it appears that 80th was their previous place and that they have since upgraded their cluster to become 77th.
    • Re:Correction? (Score:3, Informative)

      by Anonymous Coward
      Not quite, Weta have two clusters. The cluster for rent (that the article is about) was built only to supplement Weta's primary cluster (and get the films out on time).

      It has 1080 processors [top500.org] (540 blade servers) and is at rank 80.

      Weta's primary cluster (not for rent!) has 1176 processors [top500.org] and is at rank 77.

  • Doomsday scenario (Score:5, Interesting)

    by ciurana ( 2603 ) on Tuesday September 07, 2004 @03:03AM (#10174857) Homepage Journal
    Hrm...

    How soon before the bad guys set up a dummy corporation and start running nuclear bomb or protein folding simulations on this cluster? I'd be very interested (probably along with some governments) in Weta's and Gen-i screening processes. Will anyone who can foot the bill get access?

    I know, this is tinfoil hat stuff, but it's late and I get this "glass half full" visions when I'm sleep deprived.

    Cheers,

    E
    • by Vellmont ( 569020 ) on Tuesday September 07, 2004 @03:35AM (#10174991) Homepage

      How soon before the bad guys set up a dummy corporation and start running nuclear bomb or protein folding simulations on this cluster?


      The hardest part by far in making a nuclear weapon is getting the fissile material. If you are able to get highly enriched uranium you don't even need to do any simulations, the design is fairly simple and no testing is needed. Plutonium is a bit harder.

      The point though is that computer simulations of nuclear weapons is the least of your problems, and is by no means required. Computers aren't secrets, and getting a few hundred of them together in a cluster is a task anyone with $100,000 can easily accomplish. Compared to getting the required fissile material, any required computations are easy.

      I'm not sure what you're getting at with protein folding. Is their some doomsday weapon you can create by knowing how proteins fold? Even if it is, it's not a big concern. No one has gotten even close to completely simulating a protein folding. There's simply not enough computing power yet. What's been done to date are just small scale simulations.
      • anyway, I was under the impression that our nuclear simulations were for the purpose of analyzing the effects of our old-school silos. Like, this missile's been sitting around for 35 years, what will happen when we actually go and try to blow something up with it?

        Not as much using simulation to try to figure out how to make a cheaper or more deadly weapon.

    • Re:Doomsday scenario (Score:3, Interesting)

      by jesterzog ( 189797 )

      How soon before the bad guys set up a dummy corporation and start running nuclear bomb or protein folding simulations on this cluster? I'd be very interested (probably along with some governments) in Weta's and Gen-i screening processes. Will anyone who can foot the bill get access?

      I think it would depend on how open it is. The New Zealand government is strongly anti-nuclear (however rational or irrational that may be). eg. US nuclear vessels aren't allowed within the NZ economic zone. This type

    • What about all the other supercomputers [top500.org] in the world? Why are you so concerned about this particular site?
    • That does sound scarry. However, I reckon the only real plot here is how will NZ Telecom make even more money. They house Weta's cluster and they own Gen-i.

      And, yes - they are evil.

      • Goddam, where's my mod points when I need em? I still view Telecom as one of the more evil corps out there. Although it's an interesting sort of thing that they also seem to be remarkably inept in so many ways. Evil and inept. Now there's a great combo. And I don't even live in NZ currently...
  • by Splezunk ( 250168 ) on Tuesday September 07, 2004 @03:04AM (#10174861) Homepage
    my precious.....
  • Finally! (Score:4, Funny)

    by xsupergr0verx ( 758121 ) on Tuesday September 07, 2004 @03:05AM (#10174869)
    Thanks to WETA, now I can run Doom 3!
  • rent it out (Score:4, Insightful)

    by OneArmedMan ( 606657 ) on Tuesday September 07, 2004 @03:06AM (#10174876)
    to other ppl that need to render stuff. Im sure they could figure out some reasonable pricing vs CPU time etc.
    • C'mon, if you're not going to read the article, at least read the friggin' Slashdot story. If even that is too much for you, please consider at least reading the headline.

      But yeah, renting it out seems like a smart idea.
    • The best part is that this was modded "insightful".
    • You reckon? Perhaps you should ring them up, I doubt they've thought of that.
    • " to other ppl that need to render stuff. Im sure they could figure out some reasonable pricing vs CPU time etc."

      OOO! I've got an even better idea - why don't they rent it out to other companies and stuff? That way, when they're not using it, it isn't going to waste; I mean, tihnk how useful that could be.

      We should start a petition asking them to do that.
  • by Anonymous Coward
    Where can I rent an analog one?
  • An idea... (Score:5, Interesting)

    by Veridium ( 752431 ) on Tuesday September 07, 2004 @03:11AM (#10174911) Homepage
    You know, that only has ~500 nodes right? How many people out there are either out of work, or sick of doing what they're doing? Maybe we ought to get about 1000 of us nerds together in some kind of co-op, cluster our machines then rent it out? My main box is dual opterons and I already have 6 dual P-pro 200s clustered...

    Yeah, I know, the logistics of it, the devil would be in every detail... Neat to think about though.
    • Re:An idea... (Score:3, Informative)

      by mefus ( 34481 )
      Computational Grid [globus.org] is working out the logistics, and has a software package that'll get you on the grid.
      • Thanks for the link... I was thinking more of actually setting up a cluster in one location as a business co-op. It would have far greater capabilities, but at far greater costs. It was just an idea. I don't think it's feasible to get 1000 people to put up a system and a very small amount of money per person to see it happen.
    • Re:An idea... (Score:5, Insightful)

      by Vellmont ( 569020 ) on Tuesday September 07, 2004 @03:39AM (#10175010) Homepage

      You know, that only has ~500 nodes right?


      It has 500 nodes that are highly and quickly interconnected. It's like the difference between 500 people working on a problem in the same room, and 500 people spread across the country communicating by postal mail. Most interesting problems require a lot of inter-communication, so 500 slowly connect nodes isn't too usefull.
      • Yeah, I understand that a bit, I wasn't trying to put it down by saying "only" 500 nodes. It's just, pooling resources, a relatively small number of people could put together triple that number with relatively little out of pocket cost to each person. The interconnect speed wouldn't be as high, unless you got each person to purchase a blade, but dual gigabit interconnects on 1000 systems would still be a setup worthy of more than a few "interesting" problems. The trouble would be having a place to house it
        • Certainly seems like a decent idea. A super computer/cluster/whatever co-op. If you started with it widely distributed, you might make enough to take it to the single building type effort. Makes more sense and seems more likely to actually be useful and make money than any number of failed dot bombs.
          • Makes more sense and seems more likely to actually be useful and make money than any number of failed dot bombs.

            The problem with starting it out as widely distributed is the marketability of such a system. If you walk into a university or corporation trying to sell them the idea of 2000 machines run by people around the States, they'll picture half of them as being 14 year olds with modem connections.

            That might be overstating it a bit, but the way I see it, even if you have 1000 people with two machine
            • However if you talk to large corporation where the apps running on everyone's desktop are tightly controlled by the IT department and said, "Guess what, you can make money by renting your 2000 unused CPUs every night instead of running 'Flying Toasters'" they might pay attention. Especially if it was open source so they could check that you weren't secretly stealing all the data off the hard drives. You would want the client to initiate connection on port 80 so it can pierce the firewall, and you would wa
  • by Anonymous Coward
    So this thingy can stream about 1000 pr0n videos streams per blade... With 500 blades it can stream 500k streams simultaneously. This must be more than enough pr0n for everyone.
  • distcc? (Score:3, Funny)

    by SKPhoton ( 683703 ) on Tuesday September 07, 2004 @03:23AM (#10174952) Homepage
    How about using all those cpus as Distcc nodes!

    (Go Gentoo!)
  • Whack!
    Don't even say it!
  • Weta's old cluster (Score:5, Interesting)

    by SlashdotMeNow ( 799901 ) on Tuesday September 07, 2004 @04:22AM (#10175143)
    We actually bought the old cluster Jackson used (for next to nothing I might add!) It's 62 PCs (no they run Windows) totalling 124 processors (2 racks full)

    That's a total of 124GHz and 124GB of RAM. We're using it to render architectural fly-through movies.

    Hmm... I'm all hot now... Need a cold shower!
    • by SlashdotMeNow (799901) Alter Relationship
      We're using it to render architectural fly-through movies

      Give us a link to one of these movies and we would be happy to SlashdotYouNow

  • We will now truly be able to figure out how many New Zealanders it takes to change a light bulb.

  • ...otherwise my PC has 10 times more storage than a supercomputer - wow :)
    • You, and plenty of others, have failed to notice, that the stats they gave were *drumroll* PER NODE. And this machine has 502 NODES. So, multiply anything you saw by 502 (6 GB ram PER 502 NODES = 3.012 Teras of ram, 40 GB hds PER 502 NODES = 20 Teras!)

      If you can't even read the stub correctly, I know you didn't RFTA...
  • by ricky-road-flats ( 770129 ) on Tuesday September 07, 2004 @06:14AM (#10175466) Homepage
    We've had big problems using IBM blades, not least because by default they come with crappy 5400rpm laptop hard drives - and the 40 GB mentioned in these blades imply that's what they've got (which is what my predecessor ordered).

    I guess with 6 GB RAM each they shouldn't have to do much (or ANY, if I was running this) swapping, and if the jobs are tweaked to not use the hard drives too intensively, they might be OK. If what you do uses the hard drives for much, they are sh*t, to put it mildly. If you could plug these [seagate.com] into the blades, they's be very useful, quick machines. But you can't yet.

    The really crap thing is, if you do want SCSI drives in the IBM blades, you connect a module ot the side of the blade which gives you a couple of proper SCSI drive bays. Which halves the number of blades-per-bladecenter to 7.

    Given the bladecenter is 7U tall, you'd be better off with 7 1U servers with SCSI bays already in and better NIC options. The internal networking of the bladecenter is awful for everything but the simplest low-requirement setups - it's hideously expensive to give each blade a couple of gigabit connections.

    Even these cheap little things [dell.com] are 1U, take 2 U320 SCSI drives, and have dual Gigabit connections built-in.

    And I *still* can't get USB dongles to work with thes fscking blades, grumble grumble.

    Having said all that, when can I play on this thing? My Folding@Home could do with a bit of a boost, and with Hyperthreading I could have 2016 units running simultaneously.... although it might get a little warm behind the racks, 1008 2.8 GHz Xeons pump out a good bit of heat!

    • AFAIK, IBM blades are not meant for DataCenters or other I/O intensive applications. It is specially designed for SuperComputing needs and packs quite a lot of power in a very little space & provides 100% hotswap capabilities.
      For storage, the blades are provided with Fiber channel access for SANs.
    • Everyone has 1U SCSI servers, so that's no surprise.

      They provide compute density (1/2 U for two processors), nothing more. A significant portion of people I know run blades diskless off of ramdisk or NFS where decent systems actually house the storage. In blades, the only mechanical component that is not hot-swappable is the drive, so there is a desire to not rely on it at all for a lot of people I know.

      Hideously expensive to get both gigabit NICs connected? Technically speaking, those Ethernet switch
  • by Jugalator ( 259273 ) on Tuesday September 07, 2004 @07:07AM (#10175601) Journal
    Read this

    "Weta Digital Supercomputer For Hire"

    as

    "Weta Digital Supercomputer On Fire"

    Thought, whoa, finally some big news on Slashdot!

    But no... Anyone willing to go with me to put them on fire for some hot Beowulf cluster action? :-P
  • Small potatoes (Score:3, Interesting)

    by rumblin'rabbit ( 711865 ) on Tuesday September 07, 2004 @07:28AM (#10175682) Journal
    I'm surprized this is ranked as 80'th, because it's not that large by todays standards. Even some mid-sized geophysical processing companies, for example, can beat it. Large ones might have 5 or 10 times this capacity.

    If they're going to market this capacity, they had better do it quick. The shelf life of computational power is not much greater than milk.

  • by dosun88888 ( 265953 ) on Tuesday September 07, 2004 @08:29AM (#10175912)
    Do they use Tolkien Ring?

    ~D
    • Do they use Tolkien Ring?


      They tried, but they discovered an error in the protocol whereby the one who held the token wouldn't let it go.

      =)

    • You do realize most /. readers were in diapers last time anyone was using Token Ring, right? That joke probably went right over half their heads.

      And I bet the majority on here haven't had to worry about taking a whole office network down by knocking off a terminator on a 10base2...

      Damn kids today. Got everything easy.
      • o'contrarie, slashdot is teeming with old bearded grizzlies that have been around the proverbial block several times. I'm only 35 and I know what token ring is. Matter of fact, this is one of the best jokes in the thread!
    • No, but they do use elf binaries.
  • I think you should run SETI for one day. I would love to see how many work units can be computed. It would probably set a record.
  • What about #77 (Score:2, Interesting)

    Theres no reference to WETA DIGITAL's second cluster, #77 on the list. It contains 588 computers as opposed to 504.
    top 500 page for the cluser here [top500.org]
    Why don't they just combine the two. That would surely grant them a top 10 spot...
  • **angry crowd chants** we want a hobbit movie!!!

Saliva causes cancer, but only if swallowed in small amounts over a long period of time. -- George Carlin

Working...