Slashdot Mirror


TeraGrid Gets an Upgrade

The Fun Guy writes to tell us The NSF has awarded $48 million to the University of Chicago to operate and expand TeraGrid over the next five years. TeraGrid is 'a national-scale system of interconnected computers that scientists and engineers are using to solve some of their most challenging problems. TeraGrid is the world's largest open computer, storage and networking system. Only the U.S. Department of Energy's weapons laboratories have larger systems, which are dedicated to classified research.' Currently, the TeraGrid's power is just over 60 teraflops.

125 comments

  1. But by hungrygrue · · Score: 2, Funny

    Does it run Linux? Oh, I see it does... nevermind.

    1. Re:But by Anonymous Coward · · Score: 0

      Why is this funny again?

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

      Why was it funny, ever.

  2. the first thing through... by Anonymous Coward · · Score: 0

    the first thing that went through my head
    was how do i get access and how fast can it crack
    a hash.

    stupid i know, but what fun to try it.

    1. Re:the first thing through... by Anonymous Coward · · Score: 0

      pfft. Like you actualy know how to 'crack a hash'. Stop pretending, fool.

    2. Re:the first thing through... by Anonymous Coward · · Score: 0

      This "shadow" file is usually where the password hashes themselves are located..

      so what do you call an encrypted password?

      your mom?

  3. Eww more chances to refine PI by crkpot · · Score: 2, Funny

    I like it - maybe they can define PI down to an even greater degree. !!

    1. Re:Eww more chances to refine PI by CDMA_Demo · · Score: 3, Funny


      Yea, find out where is the recurring sequence of 424242....

    2. Re:Eww more chances to refine PI by Lisandro · · Score: 1

      Work ain't done until they found a circle in it!

    3. Re:Eww more chances to refine PI by Anonymous Coward · · Score: 0

      I sense that someone has read Carl Sagan's book here...

    4. Re:Eww more chances to refine PI by tehshen · · Score: 2, Informative

      42424242 is there, 242423 digits in. check it if you so desire.

      --
      Guy asked me for a quarter for a cup of coffee. So I bit him.
    5. Re:Eww more chances to refine PI by LeDopore · · Score: 0

      "Yea, find out where is the recurring sequence of 424242...."

      Funny! On a more serious note, we know Pi cannot have any repeating sequence in it, because that would make it a rational number. Pi is, however, irrational, as was proven in 1761 by Johann Heinrich Lambert (quoting wikipedia).

      --
      Expected time to finish is 1 hour and 60 minutes.
    6. Re:Eww more chances to refine PI by ToasterofDOOM · · Score: 1

      Damn! I nearly crashed my computer runnitng a regular expression on that looking for repeating parts! Note to self: uh ... don't be such a dumbass

      --
      I am Spartacus
    7. Re:Eww more chances to refine PI by Anonymous Coward · · Score: 0

      Not so, as another person showed. The correct statement is "Pi cannot have any infinitely repeating sequence in it".

  4. How does this... by Odin_Tiger · · Score: 4, Interesting

    ...stack against the likes of distributed.net and other similar projects for processing power?

    --
    Unpleasantries.
    1. Re:How does this... by darth_MALL · · Score: 0, Funny

      vertically

    2. Re:How does this... by 777v777 · · Score: 2, Informative

      As a user of teragrid, as well as other huge machines, There are some embarassingly parallel tasks like SETI at home which can be easily run on distributed systems. There are other problems where this is just out of the question. The Teragrid clusters will be much better for these types of problems.

      Tightly coupled problems just cannot be run efficiently even on clusters of workstations(COWs). It is the age old topic of using the right tool for the right job.

    3. Re:How does this... by Jeff+DeMaagd · · Score: 1

      How well does Teragrid change things? The data connection to each site is much larger, but each site has a large number of computers too, which would seem to offset the link speed.

    4. Re:How does this... by Anonymous Coward · · Score: 0

      Considering its globus powered, it probably doesn't even stack against a bunch of scientists computing equations through a 'tin cans on a string' system.

    5. Re:How does this... by Anonymous Coward · · Score: 0

      The network piece of Teragrid is almost never running at capacity from what I've observed.

    6. Re:How does this... by fshalor · · Score: 1

      Teragrid is a mixed bag... A lot of out clustered needs don't scale much beyond 16-32 nodes. And we need 16-64 GB per node to handle the data sets. And we need fast storage, since the data sets we use are huge. With Teragrid, we'd get bogged down just uploading the data.

      Just look at:
      http://access.ncsa.uiuc.edu/Releases/09.19.05_Berk eley_L.html

      I mean, the Berkeley geophysics people had to build their own. So they called on NCSA people to help build them one.

      If you have a limited budget, NSF fundig for most of your projects, and a small data set with lots of processing needed, then TeraGrid is TeraGold. .

      --
      -=fshalor ::this post not spellchecked. move along::
    7. Re:How does this... by Anonymous Coward · · Score: 0

      With Teragrid, we'd get bogged down just uploading the data.
      As of late, this is less and less true. With things like wide area parallel filesystems (IBM's gpfs has had a lot of work done, recently, in mutli-cluster mounting) and parallelized file transfers (a la gridftp), the hurdles of getting data into the sphere of teragrid, specifically, is less and less a problem. And with such large edu networks like abeliene, it's relatively easy to get a fast connection to these sites. One problem arrises, however, when you note that the machines that compose the teragrid have very little in common, as far as architecture. Commodity ia64s, Crays, SGIs... with all the variance, it's hard to do a cross-site run that's tailored to the kind of problem you want to run.

  5. I want one by stunt_penguin · · Score: 5, Funny

    but only if it comes in white....

    --
    When the posters fear their moderators, there is tyranny; when the moderators fears the posters, there is liberty.
    1. Re:I want one by daviqh · · Score: 1

      Oh, no...Everyone knows that black computers run faster!

      --
      Microsoft is like...no, it's much worse.
    2. Re:I want one by Madd+Scientist · · Score: 1

      would that be the TeraGrid U2? i don't like the red power button on it.

  6. Correction by AvitarX · · Score: 1

    TeraGrid is 'a national-scale system of interconnected computers that scientists and engineers are using to solve some of their most challenging problems.

    Replace "challenging" with "Parralell"

    --
    Wow, sent an e-mail as suggested when clicking on "use classic" banner, and got a fast response that addressed my msg
    1. Re:Correction by Anonymous Coward · · Score: 0
      Replace "challenging" with "Parralell"

      Replace "parralell" with "parallel".

  7. Imagine... by merreborn · · Score: 1, Redundant

    A beowulf cluster of these!

    1. Re:Imagine... by Danathar · · Score: 2, Insightful

      yes

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

      Redundant... brilliant irony.

    3. Re:Imagine... by Insightfill · · Score: 1

      Excellent! A mod of "redundant" to a "beowulf" joke! I hope more see and "get it".

  8. Quick, everyone! by gardyloo · · Score: 3, Funny

    A good slashdotting should be just what they want to test their servers.

  9. Correction Correction by trime · · Score: 1

    Replace "challenging" with "Parralell"

    Replace "parralell" with "parallel"
    1. Re:Correction Correction by JDevers · · Score: 1

      What did you expect from someone who calls themselves "AvitarX."

      You would think they would check the spelling before committing to a user name. Of course I think many people name their children without knowing the first thing about English linguistics and that is a rather more important decision.

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

      Replace Hominid with ad hominem ; You now have a description of your argument against the OP.
      It is pathetic that the best you could muster to counter his post is an attack on his spelling abilities.
      Please do mod this flamebait , though consider doing the same to the parent and grandparent .
      For they are doing exactly the same as I am , except with a far lesser raison d'être .

    3. Re:Correction Correction by AvitarX · · Score: 1

      I take no offence, I am too lazy to spell check my posts or learn to spell so what.

      But the name was mispelled to get an AOL screen name many years ago.

      PS. Anonymous coward, if you are going to waste your time defending/attacking people on /., do so with a name.

      --
      Wow, sent an e-mail as suggested when clicking on "use classic" banner, and got a fast response that addressed my msg
  10. The Rise of the Machine by jonfr · · Score: 2, Funny

    I wonder when the cluster discovers it's own existance and deticates that there is no need for the human race.

    1. Re:The Rise of the Machine by ikkonoishi · · Score: 1

      Just before someone plugs in an ethernet cord OF DOOM!

      Heh. Seriously if a computer gone bad ever traps me in a building, and tries to kill me off. The first thing I would do is take some power cords, cut them open, and wrap the stripped wires in them around each other. I would then go around the building, and plug those wire into any outlets I could find. The resultant short would ground the building's power eventually overloading the circuit breakers, and disabling all of the security devices that aren't on independant power.

    2. Re:The Rise of the Machine by Anonymous Coward · · Score: 0

      I, for one, welcome our new terraflopping overlords.

    3. Re:The Rise of the Machine by petermgreen · · Score: 1

      hmm all etherkillers i've seen before have had a joint in them but i can't see one in that one.

      the mains plug is moulded and i'd think it would be damn near impossible to terminate mains flex in a rj45 and also i don't see any wires running down the rj45 though its hard to tell.

      I think they just stuffed the mains flex into the boot of the rj45 without actually terminating it not really made an etherkiller.

      --
      note: i'm known as plugwash most places but i screwd up registering that here somehow in the past and now can't register
  11. They're upgrading to run... by malraid · · Score: 4, Funny

    ...Windows Vista!!

    Had to say it, sorry!

    --
    please excuse my apathy
    1. Re:They're upgrading to run... by rbanffy · · Score: 1

      Well... Windows runs the largest distributed computer in the planet.

      Too bad it only does DDOS and spam...

    2. Re:They're upgrading to run... by The-Bus · · Score: 1

      No, you're confused. Windows Vista IS a megaflop.

      --

      Small potatoes make the steak look bigger.

  12. ...in order to run Virtual Tera Patrick... by Orrin+Bloquy · · Score: 2, Funny

    ...giving a whole new meaning to Teraflops.

    --
    "Made up/misattributed quote that makes me look smart. I am on /. and I must look smart."
    1. Re:...in order to run Virtual Tera Patrick... by Anonymous Coward · · Score: 0

      ...and rimshot.

    2. Re:...in order to run Virtual Tera Patrick... by Anonymous Coward · · Score: 0

      I don't like her as much after the boobjob either, but I wouldn't call them "Teraflops"--that's just cruel.

  13. Just wait... by rewt66 · · Score: 1

    ... until I make that system my personal zombie!

    1. Re:Just wait... by ecumenical_40oz · · Score: 1

      I'd hate to see what would happen if a spammer got ahold of this thing. Imagine the shear volume of penis-enlargement emails it could send to us all... Let's hope they remember to install a firewall.

    2. Re:Just wait... by cluening · · Score: 1

      Ha! It is my personal zombie!

      --
      Posted from the wireless couch.
  14. just put xbox 360s together by alienfluid · · Score: 3, Funny

    hah .. you could just put 60 xbox 360s together to achieve that kind of power ..



    xbox 360 specs


    1. Re:just put xbox 360s together by FLAGGR · · Score: 4, Insightful

      Just incase you weren't joking.

      Microsoft can say they get x amount of tflops, and lets pretend for a moment that theoretically they are telling the truth. In reality, things *never* get to their potentials because of bottlenecks, unless of course your building super computers and have millions to invest. Microsoft and Sony can play the TFlops game, but in the end they aren't that powerful (as ArsTechnica has reported based on developer comments, they are much closer to 2-3 times the current generation in power)

      I know you were probably joking, but you were modded insigtful and I couldn't help myself.

    2. Re:just put xbox 360s together by CTho9305 · · Score: 4, Insightful

      CPU Game Math Performance
              * 9.6 billion dot product operations per second

      9.6GFLOPS*60=576GFLOPS. That's not even remotely close to 1TFLOPS, let alone 60 TFLOPS. You're off by 2 orders of magnitude.

    3. Re:just put xbox 360s together by Anonymous Coward · · Score: 0

      assuming 1000 GFLOPS = 1 TFLOP, then
      you would need approx 104 xbox 360 systems
      linked together to reach 60 TFLOP's
      So maybe just get more :)

    4. Re:just put xbox 360s together by Anonymous Coward · · Score: 0

      umm... no... you'd need 104 linked to get 1TFlop... so you'd need over 6000 to get 60 TFlops...

    5. Re:just put xbox 360s together by qbwiz · · Score: 1

      Look lower on the page - it says 1 teraflop specifically.

      --
      Ewige Blumenkraft.
    6. Re:just put xbox 360s together by Anonymous Coward · · Score: 0

      Microsoft can say they get x amount of tflops, and lets pretend for a moment that theoretically they are telling the truth. In reality, things *never* get to their potentials because of bottlenecks, unless of course your building super computers and have millions to invest.

      Although I agree with parts of this statement, and speaking from direct experience building and operating supercomputers, there are always 'bottlenecks' in a system. Some of these bottlenecks are software related, some are hardware related, but they always exist in a computer no matter how big or how much money you spend on it. The trick is to build a system that has the fewest bottlenecks for the types of applications your users will be running.

      Benchmarks show the efficiency of a machine based on a particular type or range of applications, while actual production codes will have varying efficiencies all over the map. We've run codes that were anywhere between 15 and 92 percent efficient on our system, and the statistical mean is quite low. There's no escaping that. Some of the drop off is due to the bottlenecks in the system and some is due to the code itself.

      Like a good craftsman, you don't have a single tool in your toolbox. This is why we have machines of different architectures. You find a machine that runs your code the best. This whole mentality of one machine being better than another is like saying a Phillip's head screwdriver is better than a hammer. That's ridiculous! They both have their value and their use.

  15. Translation.. by Anonymous Coward · · Score: 0
    ..for the future Computer Scientists, who are at that tender age of 10-17. You are our future, dudes. Be cool and don't do drugs!
    And so, without further ado:

    DA FUN GUY WRIETS 2 TEL US DA NSF HAS AWARDAD $48 MILION 2 TEH UNIEVRSITY OF CHICAGO 2 OP3RAET AND EXPAND T3RAGRID OV3R DA NEXT FIEV YEARS!1!11 OMG TERAGRID IS A NATIONAL-SCAEL SYSTEM OF INT3RCON3CTED COMPUT3RS TAHT SCEINTISTS AND 3NGIENERS R USNG 2 SOLVE SOMA OF THERE MOST CHALENGNG PROBLEMS11!!! WTF OMG TERAGRID IS DA WORLDS LARGEST OP3N COMPUT3R S2RAEG AND NETWORKNG SYSTEM!!!!111 WTF ONLY DA US1!1! D3PARTM3NT OF ENERGYS WEAPONS LABORA2REIS HAEV LARG3R SYSTAMS WHICH R DADICAETD 2 CLASIFEID R3SEARCH!11!1 OMG LOL CURENTLY DA T3RAGRIDS POWAR IS JUST OVER 60 TARAFLOPS1!11!1 WTF LOL

    1. Re:Translation.. by Stupendoussteve · · Score: 0

      Man, that's from this generation. Get with the times, hackers of the future will say:

      The Fun Guy writes ta tizzle us The NSF has awarded $48 million ta tha University of Chicago ta operate n expand TeraGrid over tha nizzy fizzle years . You gotta check dis shit out yo. TeraGrid is a national-scale system of interconnected cracka tizzle scientists n pimp is using ta solve some of they mizzost challeng'n problems , ya feel me?. TeraGrid is tha worlds largest open computa, storage n network'n system. Only tha U.S. Department of Energys weapons laboratizzles have pusha systems, W-H-to-tha-izzich is dedicated ta classified research . Ya fuck with us, we gots to fuck you up. Currently, tha TeraGrids powa is jizzle over 60 teraflops.

  16. Department of Energy? by RandomPrecision · · Score: 2, Insightful
    So they have computer systems larger than TeraGrid for weapons research? Imagine if the Department of Energy applied those resources to improving or replacing gasoline, supplying California's nearly-insatiable demand, creating more efficient power...

    ...you know, developing sources of energy.

    1. Re:Department of Energy? by njcoder · · Score: 3, Funny
      "Imagine if the Department of Energy applied those resources to improving or replacing gasoline, supplying California's nearly-insatiable demand, creating more efficient power..."

      Imagine if they used it to make ice cream!

    2. Re:Department of Energy? by zackeller · · Score: 1

      I imagine if they just didn't run this grid there'd be more than enough energy to go around.

    3. Re:Department of Energy? by kayak334 · · Score: 2, Insightful

      DoE currently has a 136TFLOP cluster . They use it for more than nuclear weapons research.

    4. Re:Department of Energy? by RandomPrecision · · Score: 1
      I didn't say anything about nuclear weapons, but I suppose that's what the energy department would be researching in that field.

      I'm just suggesting that perhaps weapons could be left entirely to the Department of Defense, the Department of Homeland Security, and/or the military.

      Then all of the power of that cluster would be available for energy-related research and analysis.

    5. Re:Department of Energy? by kayak334 · · Score: 1

      I'm not sure that solving the worlds energy problems has an application that would benefit from running on a cluster. Clusters are only good at certain things, obviously crunching a lot of data, sometimes data streaming in realtime. For instance, when searching for new oil, companies will place hundreds of microphones in the ocean, then detonate a small ammount of TNT and record the reflections off the ocean floor. This data needs to be processed in realtime, and many calculations need to be done on the data in order to get anything useful out of it, anda ll of this can be done in parallel. Thus, a cluster is suited perfectly to it.

      My whole point is that I wish we could just throw computing power at things and solve them quicker, but often times that isn't the case. Only certain applications lend themselves towards clustering. One of those applications is nuclear weapons research. I could be wrong, there may be fusion simulations or other such things that could benefit from running on a cluster, but in my experience I haven't seen that. Perhaps it's just a matter of where the profit is, sadly.

      IAACE (Cluster Engineer)

    6. Re:Department of Energy? by kst · · Score: 1

      That would be a great idea -- if the major bottleneck in developing new sources of energy were a shortage of computational cycles. I'm no expert on the subject, but I'd be surprised if it were.

    7. Re:Department of Energy? by bleckywelcky · · Score: 1

      One of the DoE's primary marching orders is nuclear weapons research. They need computers that large to simulate nuclear explosions to determine yields, burst effects, etc. I'd much prefer them doing this on a super computer than testing the warheads out in the south pacific.

    8. Re:Department of Energy? by slcdb · · Score: 1
      ...you know, developing sources of energy.
      But maybe Nuclear Weapons research is developing sources of energy. Perhaps the DoE's plan is to turn the entire Middle East into a solid sheet of glass and then go and steal all the oil.
      --
      Despite what EULAs say, most software is sold, not licensed.
  17. Every time I run a 500 proc batch job ... by Salis · · Score: 3, Interesting

    it makes me smile.

    It's just so ... cool.

    (And the only people who I say that to are my research group members and ... the people of Slashdot!)

    The TeraGrid is well managed too.. very few problems for such a huge system.

    --
    Favorite /. tagline: "On the eighth day, God created FORTRAN." And it was good.
    1. Re:Every time I run a 500 proc batch job ... by mccalli · · Score: 1
      I'm curious - could you please let me know what you're doing with it? And whether any special approach was needed to make best use of the system?

      In other words, suppose I threw a standard text processing job at it (trawl through 2 gigs of disparate log files, correlate, spit out unified log). Simple enough task and on the machine I use it takes about 4 minutes to complete. If I took that and ran it on the TeraGrid with no special thought to that environment (it's a single-threaded Perl script), would it run in the blink of an eye or am I still looking at a reasonable chunk.

      I'm not suggesting I'd actually do this of course - I'd optimise whatever the task was to whatever the environment is. But I'm just curious to know what it will do to bread'n'butter tasks as well as the exotics.

      Cheers,
      Ian

    2. Re:Every time I run a 500 proc batch job ... by Salis · · Score: 1

      It's an MPI environment. MPI stands for Message Passing Interface. It's a library of subroutines that allow you to send data from one processor to another. So if you wanted to process a 2 gig text processing, you could divide the text file into 500 chunks and have each processor perform some function on its chunk. Then you would collect all of the results together and save them.

      Certain programs are more 'parallelizable' than others. The programs I run (and code) are very 'embarrassingly parallel'. That means that there is very little communication between each processor while the program is running. Certain programs require a large amount of communication between processors and a lot of special attention must be given to optimizing the architecture of the parallel program.

      The bandwidth and latency between processors is pretty good, but no where near the actual speed of computing. So you need to balance the time it takes to send data from one processor to another and the time it takes to compute some results. Since the processor can do both at the same time (send/receive data while computing), it becomes a sort of art to synchronize it all together so that no processor is left idling with nothing to do.

      Hope that explains it!

      -Howard

      --
      Favorite /. tagline: "On the eighth day, God created FORTRAN." And it was good.
    3. Re:Every time I run a 500 proc batch job ... by 777v777 · · Score: 1

      As suggested by another post, you would want to parallelize your job to make it run on some number of the machines simultaneously. Using MPI could be one way of doing this. However, lets say you have an application which generates terabytes of data, and then processes it, a system like this with tons of fast storage and high bandwidth networks would be useful.

    4. Re:Every time I run a 500 proc batch job ... by vsprintf · · Score: 1

      The TeraGrid is well managed too.. very few problems for such a huge system.

      For $48 million, one should hope so. There are "national assets" of other federal agencies that don't get anywhere near that kind of funding for managing much more data. It really sounds like someone brought the bacon home from Congress.

    5. Re:Every time I run a 500 proc batch job ... by Salis · · Score: 1

      $48 million isn't much for multiple supercomputers.

      Did you look at the price tag for IBM's Blue Gene or Japan's Earth Simulator? Yeah...much more than $48 mil.

      And I can't use either of those nice government funded ones! Bastards. They won't even accept an allocation request. ;)

      "But I only want to use 15 000 processors for an hour!" Heh.

      --
      Favorite /. tagline: "On the eighth day, God created FORTRAN." And it was good.
    6. Re:Every time I run a 500 proc batch job ... by cluening · · Score: 1

      > The TeraGrid is well managed too.. very few problems for such a huge system.

      As an admin on the UC/ANL Teragrid cluster, I thank you for the compliment. Keep computin'!

      --
      Posted from the wireless couch.
    7. Re:Every time I run a 500 proc batch job ... by Colonel+Panic · · Score: 1

      How does one get an account on TeraGrid?

    8. Re:Every time I run a 500 proc batch job ... by Anonymous Coward · · Score: 0

      Ha

    9. Re:Every time I run a 500 proc batch job ... by knightri · · Score: 0

      Except when running 128 P3 700 machines using Linux equals the same performance of a single P4 3000 on XP.

      --
      'Or else pizza is going to order out for you'
    10. Re:Every time I run a 500 proc batch job ... by kst · · Score: 1

      A single-threaded Perl script would use one CPU on one node. The individual CPUs are pretty good, but only comparable to, say, a modern high-end PC.

      If you wanted to run 500 copies of your single-threaded Perl script, they'd probably all finish in about 4 minutes -- but that doesn't make very good use of the system. You could get the same results using something along the lines of SETI@Home or distributed.net.

      What makes TeraGrid special is that it's a whole bunch of CPUs (along with lots of memory, disk, and archival storage) connected by high-speed networks, with a lot of hardware and software infrastructure to make it work together. This is good for things like weather simulation, where the problem can be broken down into pieces, but the adjacent pieces need to communicate with each other.

    11. Re:Every time I run a 500 proc batch job ... by YetAnotherLogin · · Score: 2, Interesting

      If you're a member of an academic institution, you can just submit a proposal to the NSF to apply for time on the TeraGrid. I've heard at conferences that it's quite easy to get time, provided you give them a good account of the time complexity of the algorithm you want to run on their machines. And of course, that you attempt to answer a science question :-)

    12. Re:Every time I run a 500 proc batch job ... by Anonymous Coward · · Score: 1, Insightful

      The TeraGrid is well managed too.. very few problems for such a huge system.

      Actually it isn't. Each site has it's own way of doing things, its own software stack (NMI alleviates some of this), and its own particular configuration. It translates to a bunch of clusters interconnected by a high-bandwidth, low-latency network.

      Ever ran a cross-site application?

    13. Re:Every time I run a 500 proc batch job ... by Anonymous Coward · · Score: 0

      yes, yes, yes!! Listen to what this guy is saying! IMO, the TeraGrid falls quite short of the promise of grid computing.

    14. Re:Every time I run a 500 proc batch job ... by nr · · Score: 1

      Just out of curiosity, what version of Globus Toolkit are you using for TeraGrid?

    15. Re:Every time I run a 500 proc batch job ... by Anonymous Coward · · Score: 0

      iirc, we're hot on the heals of Globus 4

  18. I really hope ... by Anonymous Coward · · Score: 1, Interesting

    ... that some of that money is going to go towards securing the system. :-\

    1. Re:I really hope ... by Anonymous Coward · · Score: 0

      Ha! As an employee at one of the teragrid sites, I can tell you that that story is definitely written in true FUD fashion. There indeed are problems, but not like what they try to lead you to believe.

    2. Re:I really hope ... by kst · · Score: 1

      I wouldn't take the rawstory.com story too seriously.

      For one thing, the author seems to think that "Grid" and "TeraGrid" are the same thing. A Grid is a generic term for a set of computer resources, possibly spread across multiple administrative domains, working together using Grid software (such as Globus http://www.globus.org/). The TeraGrid is one specific Grid project.

      Beyond that, I don't know why he thinks the Department of Homeland Security has anything to do with this. The TeraGrid is not, as far as I know, used for classified research. Yes, Argonne National Laboratory is a TeraGrid site, and I'm sure they do classified work as well, but there's no reason to assume there's any connection between the two.

      The TeraGrid has the same security concerns as any other large computer system: maintaining the integrity of users' data, keeping intruders out, and so forth.

  19. Sorry to burst your bubble, but... by jd · · Score: 1

    This will only take one reactor core to power, but Vista's specs say it will require a dual-core system, minimum.

    --
    It's a small world and it smells funny; I'd buy another if it wasn't for the money; Take back what I paid (SoM)
  20. Imagine by Xarius · · Score: 1

    A Beowulf cluster of those!

    Oh, wait...

    --
    C17H21NO4
    1. Re:Imagine by Anonymous Coward · · Score: 0

      Mod parent down.

  21. Memby it will run... by dcapel · · Score: 0

    The next generation of FPS!

    --
    DYWYPI?
  22. Re:not that impressive... by FLAGGR · · Score: 1

    I bet it was still cheaper than 30 ps3's

  23. Provably non-parellelizable problems? by mosel-saar-ruwer · · Score: 1

    There are other problems where this is just out of the question... Tightly coupled problems just cannot be run efficiently even on clusters of workstations(COWs).

    If I can grab you attention for just a sec: Do you know of any books [or treatises or papers] that deal with the question of whether [some given class of] problems might be provably non-parallelizable?

    Heck, if you could just give me a few keywords to Google, I'd be really grateful.

    Thanks.

    1. Re:Provably non-parellelizable problems? by 777v777 · · Score: 2, Informative

      One paper which might help point in the right direction is "Isoefficiency: Measuring the Scalability of Parallel Algorithms and Architectures" by Grama, Gupta, Kumar. You pose a very interesting question. Any application where you have a large number of steps, each step relying upon the result from the previous step, and each step independently not parallelizable would probably fit your description. I don't know of anything off the top of my head where you couldn't parallelize some portion of it, but it is much easier to think of applications which cannot scale to large levels of parallelism. The trivial examples of good scalability like rendering frames of movies or SETI@home will scale to any cluster or set of PC's you put them on. Other things like large matrix multiplications or FFTs or N-body problems do not scale as well. In these cases as you subdivide the problem into smaller pieces for your larger number of machines, the computation on each processor will quickly become small while the communication between processors will become more significant. I guess the Alpha-Beta searches will probably not benefit by parallelization as one might imagine. You could do some proof that although you can evaluate more nodes in the game-tree, you cannot prune, and thus your search will degrade towards a parallelized MinMax search.

  24. Yeah, sure. by Musteval · · Score: 1

    The NSA certainly doesn't have computers anywhere near this. No siree.

    --
    Note to mods: I'm probably being sarcastic.
  25. What about the classified systems? by Anonymous Coward · · Score: 0

    How does this compare to the acres of systems at NSA, NRO/NIMA, CIA, and DIA?

    1. Re:What about the classified systems? by AuMatar · · Score: 1

      I could tell you, but then you'd need to be DOA.

      --
      I still have more fans than freaks. WTF is wrong with you people?
  26. Mod parent funny by kayak334 · · Score: 1

    The xbox 360 is going to use a Cell processor. The cell processor it is going to use is capable of 256GFLOPs single-precision and 25GFLOPs double-precision.

    Double-precision is all that matters in most scientific apps.

    25*60 = 1500GFLOPs = 1.5TFLOPs
    You'd need 2400 xbox360s to get to 60 TFLOPs.

    Also, Xbox360s have 512MB of RAM. This would not make for a very useful cluster node.

    1. Re:Mod parent funny by Anonymous Coward · · Score: 0

      The xbox 360 is going to use a Cell processor.

      That will be news to MS, IBM *and* Sony!

    2. Re:Mod parent funny by Ewan · · Score: 1

      The xbox 360 just uses normal powerpc procesors, not a Cell..

  27. But a systematic framework? by mosel-saar-ruwer · · Score: 1

    Other things like large matrix multiplications or FFTs or N-body problems do not scale as well. In these cases as you subdivide the problem into smaller pieces for your larger number of machines, the computation on each processor will quickly become small while the communication between processors will become more significant.

    But has anyone attempted to create a systematic [or systematizable] framework within which you might be able to prove that a certain problem was necessarily non-parallelizable?

    Cf Lagrange/Abel/Galois et al proving the non-existence of solution algorithms for fifth-order equations.

    Or Turing proving the non-existence of halting algorithms.

    1. Re:But a systematic framework? by 777v777 · · Score: 2, Interesting

      I have a non-parallelizable algorithm for you. Apply a non-associative operation to elements of an array like this:

      result = (a[0] * (a[1] * (a[2] * (a[3] *(....)))))

      Note that I use * to represent some binary operator that satisfies non-associativity. I think that this algorithm may be provably non-parallelizable, since the innermost * operation must be performed before any other * operations. Thus no two * operations can be done at the same time, and thus none of the * operations can be parallelized. Furthermore if these are the only operations being performed in the entire algorithm, then no operations in the algorithm can be parallelized. Thus the algorithm is non-parallelizable by any reasonable definition. I do assume for this proof that you cannot parallelize the * operator.

      On a side note, I could also prove that NP hard problems are still NP hard on even a large number of processors.

    2. Re:But a systematic framework? by 777v777 · · Score: 1

      Actually I meant distributive not associative...sorry

    3. Re:But a systematic framework? by Jerbiton · · Score: 1

      It has been attempted, but with modest success...

      The class of P-complete problems is widely held to be inherently sequential.

      Proving P-completeness amounts to proving that your problem is sequentially solvable in polynomial time (membership in class P), and proving that it can be reduced to another P-complete problem in polylogarithmic time on a parallel machine. Thus, you can establish that it is at least as hard to parallelize as the other P-complete problems.

      There is, to my knowledge, no formal proof that P-complete problems can't be solved efficiently in parallel, but if an efficient parallel algorithm were to emerge, a lot of researchers would be very surprised indeed.

  28. Havent we learned anything from Arnie? by andrelix · · Score: 1

    Wasnt there a movie about a computer like this? Get it up to about 100 terraflops and then throw in a little AI and next thing you know, it will only talk to you if you call it cybernet...

  29. $150 Million by Seanasy · · Score: 1

    This round of Teragrid funding is $150 Million. PSC got $52 Million. The rest is split up among the other 7 institutions. The other major partners already got big awards under earlier rounds of funding.

    Oh, and this news is a month old.

  30. Yeah, but... by superub3r · · Score: 1

    I understand It's 'powerful'... but... will it run Half-Life 2: Lost Cost?

  31. How does one get an account? by Colonel+Panic · · Score: 1

    I looked on their site and I can't see anything about getting an account (except for a form to fill-out to add users to an existing account)?

  32. Terragrid on top500.org by t35t0r · · Score: 1

    Ranked: #38, Name: TeraGrid, Itanium2 1.3/1.5 GHZ, Owner: NCSA, Country: United States, Year built: 2004, Interconnect technology: Myrinet, Number of processors: 1776, Manufacturer: IBM

    Ranked: #1 Name: BlueGene/L eServer Blue Gene Solution, Owner: DOE/NNSA/LLNL, Country: United States, Year built: 2005, Number of processors: 65536, Manufacturer: IBM.

    Is it kind of wierd that in /usr/include/limits.h:

    /* Maximum value an `unsigned short int' can hold. (Minimum is 0.) */
    # define USHRT_MAX 65535

    and the number of processors on that thing is 65536 (add a 0th processor). Anyone who knows about HPC systems care to elaborate how that happened?

    1. Re:Terragrid on top500.org by zopf · · Score: 1

      I'm confused as to what the problem is here... I suppose each processor has an integer ID, ranging from 0-65535 (for a total of 65536 ID numbers). It seems like they simply filled the system to the maximum.

      --
      Did you see the pool? They flipped the bitch!
    2. Re:Terragrid on top500.org by 777v777 · · Score: 1

      That is a convenient number if you look at the bluegene configuration. Bluegene/L at LLNL has 64 racks, each with 1024 processors. 64*1024=65536. or maybe this is the 32 racks with 2048 processors. The counting of these things is ambiguous, but Each rack has 1024 Nodes(each node having two processors). And can be used in two modes, a coprocessor mode where one just does network stuff. All this information is public, so you can search for it on Google. It is just a happy power of two that you found.

  33. Once that happens by commodoresloat · · Score: 1

    The TeraGrids have already won.

  34. Re:Teragrid on top500.org by kst · · Score: 1

    USHRT_MAX has to be one less than a power of two on any system that uses binary integers (i.e., on just about any system in the real world); 65535 is the minimum allowed value. It's not a significant limitation; if you need bigger numbers, use a bigger type.

    For various reasons, the number of nodes in a system is often conveniently some power of 2.

    The fact that USHRT_MAX+1 and the number of nodes in the BlueGene/L system happen to be the same power of 2 is purely coincidental. It's conceivable, I suppose, that something is using a variable of type short as a node index, but I doubt it.

  35. Terminator by Danathar · · Score: 1

    Sombody make sure Cyberdyne systems was NOT one of the vendors involved......

    1. Re:Terminator by Anonymous Coward · · Score: 0

      As a former UChicago maroon, I can assure you that Cyberdyne are mere puppies in comparison to the evil that lurks around Hyde Park.

  36. Dept. of Energy? Weapons Computers?? by meador · · Score: 1

    In case you are wonder what the DOE needs teraflop weapons computers for...

    think ICBM, baby... http://www.sandia.gov/media/online.htm

  37. Funny because of repetitive nature of the comment by jurt1235 · · Score: 1

    Sometimes the question "Does it run linux" is asked on a relevant subject, and is not funny, but just inquisitive. It becomes humor when it is used on totally unrelated subjects. It is the same way comedians make certain jokes, or put a theme in a show.

    In this case the comment is perceived as funny, on the next it can be perceived as redundant, troll or overrated.

    --

    My wife's sketchblog Blob[p]: Gastrono-me
  38. Deadlocks by commanderfoxtrot · · Score: 1

    And you'd then be completely screwed as a serious security door will have deadlocks to stop you doing just that!

    Just don't fry the comms kit so someone outside can hopefully come and find you...

    --
    http://blog.grcm.net/
  39. Sounds Familiar by lbmouse · · Score: 1

    Hmmm, TerraGrid... can anyone say 'SkyNet'?