Slashdot Mirror

← Back to Stories (view on slashdot.org)

Supercomputer Designer Asked To Improve Robo-Bugs

Posted by Soulskill on from the can-you-make-them-super-robo-bugs? dept.

Nerval's Lobster writes "The man who designed the world's most energy-efficient supercomputer in 2011 has taken on a new task: improving how robo-bugs fly. Wu-chun Feng, an associate professor of computer science in the College of Engineering at Virginia Tech, previously built Green Destiny, a 240-node supercomputer that consumed 3.2 kilowatts of power—the equivalent of a couple of hair dryers. That was before the Green500, a list that Feng and his team began compiling in 2005, which ranks the world's fastest supercomputers by performance per watt. On Feb. 5, the Air Force's Office of Scientific Research announced it had awarded Feng $3.5 million over three years, plus an option to add $2.5 million funding over an additional two years. The contract's goal: speed up how quickly a supercomputer can simulate the computational fluid dynamics of micro-air vehicles (MAVs), or unmanned aerial vehicles. MAVs can be as small as about five inches, with an aircraft close to insect size expected in the near future. While the robo-bugs can obviously be used for military purposes, they could also serve as scouts in rescue operations."

21 comments

  1. Sci-fi by Nethemas+the+Great · · Score: 2

    You know, it's sad, fascinating and scary to witness sci-fi come to be reality. It looks like we need to start working on counter-measures to these little (micro) intrusion devices.

    --
    Two of my imaginary friends reproduced once ... with negative results.
  2. From my experience by GameboyRMH · · Score: 1

    See if there are any motor-sailplane control algorithms you can copy, and don't fly them unless the wind is very still.

    --
    "When information is power, privacy is freedom" - Jah-Wren Ryel
  3. I'm thinking... by Dngrsone · · Score: 2

    The paparazi will have them working well before the military does.

  4. Hardware or Software by schneidafunk · · Score: 1

    I wonder which is more important in this situation, an improvement in the software or hardware.

    --
    Some people die at 25 and aren't buried until 75. -Benjamin Franklin
  5. Clusters NOT super computers by colin_faber · · Score: 1

    Sorry but a multi-megaflop machine is NOT a super computer, not even in the most basic stretch of the word. Yes it's impressive how much compute these guys can get out of so little power, but lets not kid our selves here. These machines are in no way super computers as commonly defined.

    1. Re:Clusters NOT super computers by LordByronStyrofoam · · Score: 1

      One very useful definition of a supercomputer is "any computer with performance within an order of magnitude of the fastest computer on the TOP500 list".

      --
      Slashdot's name? When my compiler sees /. it generates a warning about a badly formed comment.
  6. Not so fast by schneidafunk · · Score: 1

    I think you meant multi-gigaflop:

    " the top spots on the list have been taken over by machines that combine commodity processors with coprocessors or graphics processing units (GPUs) to form heterogeneous high-performance computing systems.

    With all eyes on the new TOP500 number one system, Oak Ridge National Labs' Titan, it was a system belonging to a neighbor at the University of Tennessee that debuted at the top of the November Green500 List. The National Institute for Computational Sciences' Beacon system has set the new energy efficiency bar at nearly two-and-a-half billion floating-point operations per second (gigaflops) per watt. Employing Intel's Sandy Bridge series of Xeon central processing units (CPUs) and four of Intel's Xeon Phi coprocessors per node Beacon achieved a peak 112,200 gigaflops of performance running the LINPACK benchmark while consuming only 44.89 kW of power."

    source here

    --
    Some people die at 25 and aren't buried until 75. -Benjamin Franklin
    1. Re:Not so fast by colin_faber · · Score: 1

      Hi Yes,

      But this is comparing a fast bicyclist to a fast space craft in the sense of speed difference here.

      We're talking machines that are pushing into the tens and hundreds of petaflops (next generation).

      I'm fine with calling them low-power high performance clusters, but calling them super computers is something completely different altogether.

    2. Re:Not so fast by schneidafunk · · Score: 1

      Fair enough and you are correct. I just looked it up and the 2012 winner for fastest supercomputer was the Cray Titan at 17.59 PFLOPS.

      source here

      --
      Some people die at 25 and aren't buried until 75. -Benjamin Franklin
    3. Re:Not so fast by Anonymous Coward · · Score: 1

      TL;DR The term supercomputer is not very precise but it really doesn't take much to build one that would be on the Top500 list (which one could argue is "a definition" of what a supercomputer is.)

      While I would argue that just because something doesn't make the Top500 doesn't mean it isn't a supercomputer, no one can contest that being on the Top500 list makes your machine a supercomputer. Hence a better working definition may be to look at the BOTTOM machine on the Top500. As of November 2012, that would be a machine with a maximum computational capacity of 76.4 TFlops (out of 137.2 TFlops peak) using 13,560 cores.

      Next, even Cray produces clusters as supercomputers. Hence just because it is a cluster does not mean it is *not* a supercomputer.

      Just to muddy the waters further, I have a GPU in my desktop that achieves nearly 1/2 TFlop peak when used for computations. With a computational efficiency of about 50% (about right for some machines on the Top500), I would need 175 of them networked together to approximate the speed of the lowest machine. Could I afford 175 of them? Perhaps not but a company could. (I am neglecting the network effects which become challenging for distributed systems. But you get the idea.) The real reason why a company wouldn't buy one but rent time on someone else's machine is because some domain's problem sizes are growing so fast that the machine will be obsolete before is is out of maintenance.

      Note: IAASD ("I am a supercomputer designer") as part of my day job. A machine I had a major hand in designing (with around 5000 cores) is still on the list. (Go ahead and dismiss me for being an AC. It won't make you any more right nor me any more wrong.)

    4. Re:Not so fast by Anonymous Coward · · Score: 0

      Interestingly enough, Cray just purchased Appro, Beacon's manufacturer. So technically Cray can now lay claim to the #1 spot on TOP500 and Green500.

  7. missed the point by schneidafunk · · Score: 1

    I'm not really sure what you are trying to say here, because the only thing he's going to be working on is improving computer power, either through hardware or software. All he is doing is a simulation, as stated in the summary, his contract is to: " speed up how quickly a supercomputer can simulate the computational fluid dynamics of micro-air vehicles."

    --
    Some people die at 25 and aren't buried until 75. -Benjamin Franklin
  8. correction needed by Anonymous Coward · · Score: 0

    Wu-chun Feng, an associate professor of computer science in the College of Engineering at Virginia Tech

    When I went to VA Tech, the CS program was where it's supposed to be, in the Mathematics department, which was in the College of Liberal Arts. Eventually, a marketing decision pushed Software Engineering, a new program within CS, into the College of Engineering. Now it looks like marketing pushed all of CS into the Engineering College. Disney World isn't the only place you'll find imagineers... they're also at VA Tech! Computer Scientists imagining that they're actually engineering something!

    1. Re:correction needed by Anonymous Coward · · Score: 0

      It comes down to simple dollars and donuts. The Math department has no monies, the engineering department has lots. If they wanted to suckle on that sweet sweet nipple they had to change allegiances. Not saying it's right just saying that's the reality

  9. Take that! by Anonymous Coward · · Score: 0

    Suck it Georgia Tech!!~

  10. Fly Paper by Anonymous Coward · · Score: 0

    They don't sell as much fly paper as they used to. I'm expecting a boom in sales!

  11. Counter-meansures by Su27K · · Score: 1

    I think Bill Gates is already working on it.

  12. Not so fast by Anonymous Coward · · Score: 0

    These machines are in no way super computers as commonly defined.

    [b][url=http://www.bangliwj.net/]Glass Door Lock[/url][/b],[b][url=http://www.bangliwj.net/]shower hinge[/url][/b],[b][url=http://www.bangliwj.net/]floor spring[/url][/b]
    [b][url=http://www.kenill-tape.com/]double sided tape[/url][/b],[b][url=http://www.kenill-tape.com/]masking tape[/url][/b],[b][url=http://www.kenill-tape.com/]packing tape[/url][/b]

  13. Not so fast by Anonymous Coward · · Score: 0

    These machines are in no way super computers as commonly defined.

    floor spring,shower hinge,Glass Door Lock
    packing tape,double sided tape,masking tape