back to article King Kong's dangling cluster cut loose

The supercomputer used to animate Peter Jackson's 2005 version of King Kong has been donated to New Zealand's Auckland University of Technology (AUT). The supercomputer comprises 200 IBM HS20 Blade servers, a piece of kit Big Blue sold back in 2005 and a couple of preceding years. IBM's page for the HS20 suggests the machines …


This topic is closed for new posts.
  1. frank ly

    I'm wondering .....

    .... how much it would cost to run, maintain and service it.

    1. Gray Ham

      Re: I'm wondering .....

      One would hope it would cost a bit less to maintain and run than actually sourcing a brand new one. Still, as my grandma used to say, "don't look a gift horse in the mouth".

      1. Captain Scarlet Silver badge

        Re: I'm wondering .....

        200 IBM HS20 Blade servers, so thats 17 Blade Centres.

        I doubt they would have much computing power these days but I suppose it can be used as a way to experiement with clustering. Although I can't see anywhere about storage or whether they are getting the rear add-ons for the Blade Centres.

      2. JLH

        Re: I'm wondering .....

        "One would hope it would cost a bit less to maintain and run than actually sourcing a brand new one."

        Well... maybe not.

        As machines become older, it gets harder to source the parts - particularly DIMMs and CPUs.

        So HPC manufacturers ask increasing amounts to keep older machines under maintenance.

        Of course it is in the interests of manufacturers to sell you new machines.

        In my experience it is the DIMMs which will go faulty most often on a machine such as this.

    2. I. Aproveofitspendingonspecificprojects

      Re: I'm wondering .....

      "how much it would cost to run, maintain and service it."

      However much it cost and even if it was made of gold it would still be cheaper to give it away -which he did, just in time to prevent Microsoft suing him for patent infringement next time he made a film.


      We need a pirate icon with a cross over it for posts that refer to thwating 2@everying progressive thwartters.)

  2. Cuddles Silver badge

    1.4 TFLOPS?

    No wonder they're just giving it away, that's less than a new console, let alone a decent workstation PC. More memory and scope for playing with parallelisation, but certainly not something anyone's going to pay actual money for these days.

  3. jabuzz


    My guess is that the real problem with a render farm hand off is the interconnect. Rendering frames for a movie is trivially parallel; each frame can be rendered utterly independent of any other frame, so while a node is rendering a frame it has no requirement to communicate with any other node, till it gets to the end when it needs to upload the frame to a central storage area. You can happily do this with gigabit ethernet.

    However for a real HPC cluster that would not cut the mustard and you need high speed low latency interconnect, either Infiniband or something similar. Retrofitting this into an old system is simple madness.

  4. Captain Hogwash Silver badge

    Re: King Kong's dangling cluster cut loose

    See illustration.

  5. Pirate Dave Silver badge


    More important than "will it run Crysis" is "how many BitCoin hashes can it generate per hour?"

This topic is closed for new posts.

Biting the hand that feeds IT © 1998–2019