Slashdot Mirror


More Details Of IBM's Blue Gene/L

Bob Plankers writes "By now we've all heard about IBM's Blue Gene/L, LLNL's remarkable new supercomputer which is intended to be the fastest supercomputer on Earth when done (360 TeraFLOPS). IBM has released some new photos of the prototype, and renditions of the final cluster. Note that the racks are angled in order to permit hot air to escape vertically and reduce the need for powered cooling. The machine uses custom CPUs with dual PowerPC 440 processing cores, four FPUs (two per core), five network controllers, 4 MB of DRAM, and a memory controller onboard. The prototype has 512 CPUs running at 700 MHz, and when finished the entire machine will have 65536 dual-core CPUs running at 1 GHz or more. Stephen Shankland's ZDnet article also mentions that the system runs Linux, but not on everything: 'Linux actually resides on only a comparatively small number of processors; the bulk of the chips run a stripped-down operating system that lets it carry out the instructions of the Linux nodes.'"

119 comments

  1. fp by Anonymous Coward · · Score: -1, Offtopic

    f6risth p4osthle

    1. Re:fp by Anonymous Coward · · Score: -1, Offtopic

      yeah, I got first post. I have the true will of warrior you cock-smoking teabaggers. w00t!

  2. Convicted corporations rewarded not punished! by Anonymous Coward · · Score: -1, Offtopic

    Convicted Corporations Receive Perks Instead of Punishment

    Sources:
    ASHEVILLE GLOBAL REPORT, No. 183, July 18-24, 2002
    Title: "Corrupt Corporations Still at Work in Developing World"
    Author: Emad Mekay

    MOTHER JONES, May/June 2002
    Title: "Unjust Rewards"
    Author: Ken Silverstein

    Faculty Evaluator: Laurie Dawson Ph.D., Diana Grant Ph.D.
    Student Researcher: Lindsey Brage, Terri Freedman

    American energy giant, Enron, and telecommunications company, WorldCom, committed massive corporate fraud and illegal acts. Declaring bankruptcy in December 2001, they left thousands of American workers jobless and without pensions. The Institute for Policy Studies in Washington D.C. and Corpwatch, a multinational watchdog group, has uncovered evidence of bribery scandals, environmental degradation and violations of international and labor laws. Yet Enron still has 25% interest in a Bolivian oil company called Transredes. Working with Shell Oil, the company is building a pipeline through Bolivia's Chaco Forest region, an area internationally known for its biodiversity, endangered species and the ancestral homeland of the indigenous Guarani and Guianeck peoples. In December 2002, Transredes was granted $220 million in loans from the International Development Bank, to be backed by U.S. taxpayer dollars.

    Enron was also responsible for cutting down the last intact, dry tropical forest in the world, Bolivia's 15-million acre Chiquitano Forest, for another gas pipeline. The Chiquitano Forest was home to the endangered marsh deer, maned wolf, jaguar, ocelot and the hyacinth macaw. The World Wildlife Fund ranked the area one of the world's 200 most endangered eco-regions. The Overseas Private Investment Corporation (OPIC), during the Clinton administration in 1999, approved loans for Enron's pipeline, which could have skirted the forest at an additional expense to the company. Officially, OPIC is mandated to protect ecologically sensitive areas.

    WorldCom still profits from its extensive telephone and internet networks throughout Latin American, Asia, Europe and Africa. Enron has additional business interests throughout Central and South America. In northwest India a power plant, which they co-own with Bechtel and General Electric, is so controversial that Enron officials face threats of being arrested on the spot if they enter the country. According to Nadia Martinez of the Institute for Policy Studies, "Enron and WorldCom are just symptoms of the way companies are able to do business without accountability."

    In 2000, Clinton issued an order that provide clear guidelines regarding the awarding of federal contracts. The new "contractor responsibility rule" specified that federal officers should weigh "evidence of repeated, pervasive or significant violations of the law," for example cheating on prior contracts, violating environmental and safety laws, labor rights, consumer protection laws, and antitrust activities. President Bush quietly killed the rule requiring officials to ban federal contractors with a record of violations of workplace safety and other laws.

    The Congressional Research Service issued an opinion concluding that the secret suspension of the rule (no issuing of public notice or soliciting of comments) was probably illegal, but the move went virtually unreported in the media. The government does not maintain a central database to store information on contractors' records of compliance with the law. The EPA and OSHA maintain their own lists of corporate violations, but parent companies are not always linked to their subsidiaries. "There's no process built into the review system," says Gary Bass, executive director of OMB watch, a Washington-based advocate of government accountability.

    A six-month investigation by Mother Jones of the nation's 200 largest contractors found that the government continues to award lucrative contracts to dozens of companies that it has repeatedly cited for serious workplace and environmental violations. Among the findin

  3. Red Lips by Anonymous Coward · · Score: -1, Offtopic

    He walked down the dark stone corridor of the castle's underground chambers. The torch in his hand was the only light source. It was cold and damp. He shivered with cold and fear. He had to find her and kill her before the sun had set. He had barely an hour before it set now.

    He came upon the heavy wooden and steel door. It was locked from the inside. He knew that her rest chamber must be on the other side. He drew a heavy steel chisel from his belt and drove it into the wood around the lock with the heavy hammer he carried. The wood was hard and it required a tremendous effort to break away such a small piece of the door. He had to work faster or he would be doomed.

    As he worked on the door, he remembered the first time he had seen her. The dark mysterious woman was so beautiful. Her heavy black satin cloak with a slight trimming of red outlining her face had hidden most of her from view. But those eyes and those lips were not hidden. Such deep, dark, beautiful eyes. They stole a man's very soul as he looked at them. And her mouth. Those full, deep red, luscious lips. The way they pouted and begged to be kissed. It was no wonder no man had resisted her. She had said only, "Come to me," to his friend that night. His friend had left with her a moment later amid catcalls from all the others about getting laid. They had never seen him again, alive. And the others, there had been six more. He knew. That was why he was here to stop her.

    The lock was barely holding now. He had a mere twenty minutes before sunset. He drew back the hammer and hit the door with all his might. The wood splintered, the lock skidded across the stones at his feet and the door opened a few inches. He pushed the door open further and found a lighted chamber on the other side.

    The walls were draped in flowing coverings of red satin. Torches burned brightly in several spots. The floor was covered with an ancient Persian rug. In the center of the floor was a long wooden chest.

    He moved to the chest and inspected it. There were no locks to break. He took hold of the lid and raised it. It was heavy but his strong arms could easily handle it. It fell open on the other side of the chest. Inside was an oblong bundle wrapped in black silk. Each of the four sides of the chest were on hinges with releases on the inside. He unlatched the releases and the sides fell open. The bundle was now fully accessible and resting on a cushion covered in the same black silk. He raised up the hammed and drew the long oak stake from his belt. He placed the point against the center of the bundle and prepared to drive deep into the bundle. But he couldn't. He had to be sure he struck the heart or he would be doomed. The only way was to draw back the silk covering.

    He mustered his courage and strength. He grabbed hold of the soft sensuous material and pulled it away. The sound of the silk slipping against itself was soothing and yet, unsettling at the same time. Each of the folds of material now slid from the form it covered as if a package was unwrapping itself. When the last fold fell away, she lay uncovered before him.

    He stared at her. He had never seen her like this before. Never without her black cloak. She lay sleeping. She was so very beautiful. Her lips were full and red, as if painted with blood. Her skin was soft and glowing. She had long, very dark, curly, flowing hair that looked like delicate silk threads. Her breasts were huge and voluptuous, bursting from her gown with its low cut front. Her breasts called to him and invited his caresses. Her dress was red satin and form fitting. Her figure was so enticing with its well defined curves. There was a generous slit up side of the dress that left one leg in plane view. Her legs were long and shapely. They were so sexy in their black lace stockings that ended at mid-thigh. Her black leather shoes with their four inch spike heels were the perfect end to her alluring legs. The contrast of her white skin, black stockings and red dress was hypnotic.

    He checked hi

    1. Re:Red Lips by Anonymous Coward · · Score: -1, Offtopic

      That was quite impressive. I mean, Really.

      Why was I here again? Oh .. yeah, something about IBM.

      Oh whatever.

      Write that yourself?

      if not, where can I get more?

    2. Re:Red Lips by Anonymous Coward · · Score: -1, Offtopic

      Try this.

  4. Doom3? by arcanumas · · Score: 5, Funny

    Well, it may be able to play Doom3 when it is released.

    --
    Slashdot Sig. version 0.1alpha. Use at your own risk.
    1. Re:Doom3? by Anonymous Coward · · Score: -1, Flamebait

      Do you know something the rest of the universe doesn't?

      AFAIK, Doom3 runs on normal PC hardware.

      What you're saying is like saying "It may be able to play Quake3 when it comes out" before Quake3 came out. Quake3 runs damn good on my 450 Mhz PC... so I guess you're just a moron.

    2. Re:Doom3? by armando_wall · · Score: -1


      Just what I needed.

      Yet another machine that will beat me in GNUChess!

    3. Re:Doom3? by PingPongBoy · · Score: 1

      Well, it may be able to play Doom3

      This time guess who/what will hold the joystick

      --
      Know your pads. One time pad: good for cryptography. Two timing pad: where to take your mistress.
  5. Why GWB's administration is a threat to peace by Anonymous Coward · · Score: -1, Offtopic

    The Neoconservative Plan for Global Dominance

    Sources:
    The Sunday Herald
    September 15, 2002
    Title: "Bush Planned Iraq 'regime change' before becoming President"
    Author: Neil Mackay

    Harper's Magazine
    October 2002
    Title: "Dick Cheney's Song of America"
    Author: David Armstrong

    Mother Jones
    March 2003
    Title: "The 30 Year Itch"
    Author: Robert Dreyfuss

    Pilger.com
    December 12, 2002
    Title: "Hidden Agendas"
    Author: John Pilger

    Random Lengths News
    October 4, 2002
    Title: "Iraq Attack-The Aims and Origins of Bush's Plans"
    Author: Paul Rosenberg

    Project Censored wishes to acknowledge that Jim Lobe, the Washington, D.C. correspondent for Inter Press Service (IPS), has been covering the ways in which neo-conservatives, using the Project for the New American Century (PNAC) among other mechanisms, used the 9/11 attacks to pursue their own agenda of global dominance and reshaping the Middle East virtually from the outset of the Bush administration's "war on terrorism." For more information, please vist the following link: http://www.ipsnews.net/focus/neo-cons/index.asp

    Faculty Evaluators: Phil Beard Ph.D. and Tom Lough Ph.D.
    Student Researcher: Dylan Citrin Cummins

    Corporate Media Partial Coverage:
    Atlantic Journal Constitution, 9/29/02, The President's Real goal in Iraq, By Jay Bookman

    Over the last year corporate media have made much of Saddam Hussein and his stockpile of weapons of mass destruction. Rarely did the press or, especially, television address the possibility that larger strategies might also have driven the decision to invade Iraq. Broad political strategies regarding foreign policy do indeed exist and are part of the public record. The following is a summary of the current strategies that have formed over the last 30 years; strategies that eclipse the pursuit of oil and that preceded Hussein's rise to power:

    In the 1970s, the United States and the Middle East were embroiled in a tug-of-war over oil. At the time, American military presence in the Gulf was fairly insignificant and the prospect of seizing control of Arab oil fields by force was pretty unattainable. Still, the idea of this level of dominance was very attractive to a group of hard-line, pro-military Washington insiders that included both Democrats and Republicans. Eventually labeled "neoconservatives," this circle of influential strategists played important roles in the Defense Departments of Ford, Reagan and Bush Sr., at conservative think tanks throughout the '80s and '90s, and today occupies several key posts in the White House, Pentagon, and State Department. Most principal among them are:

    Dick Cheney and Donald Rumsfeld, our current Vice-President and Defense Secretary respectively, who have been closely aligned since they served with the Ford administration in the 1970s;
    Deputy Defense Secretary Paul Wolfowitz, the key architect of the post-war reconstruction of Iraq;
    Richard Perle, past-chairman and still-member of the Pentagon's Defense Policy Board that has great influence over foreign military policies;
    William Kristol, editor of the Weekly Standard and founder of the powerful, neo-conservative think-tank, Project for a New American Century.

    In the 1970s, however, neither high-level politicos, nor the American people, shared the priorities of this small group of military strategists. In 1979 the Shah of Iran fell and U.S. political sway in the region was greatly jeopardized. In 1980, the Carter Doctrine declared the Gulf "a zone of U.S. influence." It warned (especially the Soviets) that any attempt to gain control of the Persian Gulf region would be regarded as an assault on the vital interests of the U.S. and repelled by any means necessary, including military force. This was followed by the creation of the Rapid Deployment Force -- a military program specifically designed to rush several thousand U.S. troops to the Gulf on short notice.

    Under President Reagan, the Rapid Deployment For

  6. it's all cool and everything... by Dreadlord · · Score: -1, Funny

    ...but can this computer actually run:

    10 LET x = 1
    20 LET y = 2
    30 PRINT x + y

    Now this would be impressing

    --
    The IT section color scheme sucks.
    1. Re:it's all cool and everything... by krumms · · Score: 2, Funny

      ...but can this computer actually run:

      10 LET x = 1
      20 LET y = 2
      30 PRINT x + y


      This seems to be a "does it run Linux?" joke gone horribly wrong.

      Or is that kernel code you're posting? :P

    2. Re:it's all cool and everything... by Uerige · · Score: 0, Funny

      Or is that kernel code you're posting? :P
      I sure hope he's paid the royalties to SCO, then.

    3. Re:it's all cool and everything... by Dreadlord · · Score: 1
      yep, it's actually supposed to get something like:

      nope, it runs Linux!

      As a reply ;)

      --
      The IT section color scheme sucks.
    4. Re:it's all cool and everything... by Anonymous Coward · · Score: 0

      Yea, I just hope it won't do:

      PRINT X+X
      or
      PRINT Y+Y

      and come out with a meaningful answer. Ick! That would just be wrong. But then again, I hope that it will never run a BASIC interpreter either

    5. Re:it's all cool and everything... by gnu-generation-one · · Score: -1, Offtopic

      42 is the answer. what is the question?

      Why do people like to drink tea?

      Play a game of noughts and crosses against yourself

      Lipread from a video camera

      This is an XBox, what is the software signing key?

    6. Re:it's all cool and everything... by Anonymous Coward · · Score: 0

      This has nothing to do with Linux. This joke is much more crass. Think about it. Genes make chromozones. What happens when you put X and Y chromozones together. (Hint: storks with packages) Now, if a computer really could do that it would be amazing -- and slightly unsettling.

    7. Re:it's all cool and everything... by gazbo · · Score: 4, Funny

      Yes. The thought of a creature with two X chromosomes is horrifying.

    8. Re:it's all cool and everything... by 42forty-two42 · · Score: 2, Funny

      Don't be silly, kernel code would be:

      #include &ltlinux/config.h>
      #include &ltlinux/module.h>
      #include &ltlinux/kernel.h>
      #include &ltlinux/types.h>

      MODULE_LICENCE("GPL");

      __asmlinkage inline unsigned int add_x_plus_y(unsigned int x, unsigned int y){
      unsigned int ret;
      spin_lock_irq(&current->arith->lock);
      current->arith->accum = x;
      current->arith->oprand = y;
      __perform_add(&current->arith);
      ret = current->arith->accum;
      spin_unlock_irq(&current->arith->lock);
      return ret;
      }

  7. Turing machine will be Turing machine by incal · · Score: 3, Interesting

    no matter how many cpu's it will get. Maybe its better to invest time and resources in the David Deutsch research of quantum machines? http://www.qubit.org/people/david/David.html

    1. Re:Turing machine will be Turing machine by IWannaBeAnAC · · Score: 2, Insightful
      And what gave you the impression that this research was not being done?

      On the other hand, it is nice to have a fast computer to play with now, not in 50 years time!

    2. Re:Turing machine will be Turing machine by DAldredge · · Score: 1

      Well, how much money and time have you donated to David Deutsch's project?

    3. Re:Turing machine will be Turing machine by iggymanz · · Score: 1

      this thing isn't even up to what a Turing machine could do, it doesn't have infinite memory.

      I have a turing machine but I can't find the end of the tape to thread it up.

  8. Infinite by Raynach · · Score: 5, Funny

    I'm really impressed with this computer. I think it's going to be the first computer that can finish an infinite loop in under an hour.

    --
    - A
    1. Re:Infinite by mOoZik · · Score: -1, Redundant

      If an infinite loop is infinite, how can it be finished?

    2. Re:Infinite by Bronster · · Score: 5, Funny

      If an infinite loop is infinite, how can it be finished?

      That would be the sound of the joke wooshing around, and around, and around, and ... your head.

    3. Re:Infinite by Anonymous Coward · · Score: 0

      In case of an infinite loop, the machine goes down in flames in an hour, thus effectively finishing the loop.

    4. Re:Infinite by mOoZik · · Score: -1, Flamebait

      Ah...it takes someone like yourself to realize a joke as lame as that. ;)

    5. Re:Infinite by Anonymous Coward · · Score: 0

      Don't worry about it - the geek sense of humour is typically atrocious. They rarely innovate, only copy tired cliches (a bit like most Linux development ..)

    6. Re:Infinite by devnullify · · Score: 0

      You just had to make your own post prove your point didn't you? :P

    7. Re:Infinite by Anonymous Coward · · Score: 0

      I think it's going to be the first computer that can finish an infinite loop in under an hour.

      but it runs linux! surely you are aware of the quote:

      "We all know Linux is great...it does infinite loops in 5 seconds."
      -- Linus Torvalds

    8. Re:Infinite by Anonymous Coward · · Score: 0

      Not playing with a full deck of cards, are we?

    9. Re:Infinite by Anonymous Coward · · Score: 0

      And someone even more lame and fucking stupid not to realize it.

      Glad to see that you know your place!

    10. Re:Infinite by vegetasaiyajin · · Score: 1

      If an infinite loop is infinite, how can it be finished?
      You only have to finish the first iteration in half a second. The second in 1/4th of a second. The third in 1/8th of a second. In general, the nth iteration in half the time required for the n-1th iteration.

      After one second the loop will be finished.

      --

      My heart is pure, but make no mistake, it's pure evil
    11. Re:Infinite by md65536 · · Score: 1

      Urghhh... infinity paradox!

      I don't think the loop will be finished after 1 second. An infinite loop is "One that never terminates". There can be no satisfiable loop exit condition in a finite amount of time (interpretting "never" to mean "not at any time; not while forever is still happening").

      Your machine would require being able to do an infinite number of loops in an infinitely small time. You could also do it if you had a machine that could do an infinite number of iterations simultaneously in finite time, which I personally think would be easier to build. After 1 second, an infinite number of iterations will have taken place, but that doesn't necessarily mean that it's "finished". You can complete an infinite number of iterations and still have an infinite (or infinity times a billion!) more to go.

      What blows my mind is... what happens to your machine at 1 second? It's doing an infinite number of iterations in an infinitely small time. Does it go on like this? Does it keep going faster? Does it destroy the universe?

      If you interpret "never" as "not after any finite number of iterations", then maybe you can get it to finish, depending on the loop condition.

  9. At least by Anonymous Coward · · Score: 0

    ... the title says its a dupe.

    Way to go /.!!

  10. Travelling salesmen. by Manywele · · Score: 5, Funny

    This will be sure to boost the effeciency of travelling salesmen everywhere.

    1. Re:Travelling salesmen. by Vegard · · Score: 5, Funny

      Actually, we better let the salesmen travel. It's a little known secret that the reason computers are so bad at solving the travelling salesmen problem is that those who design computers are technicians, and everyone knows that tech people hates salesmen, so the longer they spend travelling, the better for the techs.

    2. Re:Travelling salesmen. by Anonymous Coward · · Score: 0

      That was so not funny, it's pathetic.

      Go mods, for thinking it was funny and modding it up. Feh.

    3. Re:Travelling salesmen. by afidel · · Score: 1

      The traveling salesman problem should really be renamed to the airline scheduling problem. Airlines spend a LOT of money on large computers to ring more efficiency out of their huge capital investments.

      --
      There are 4 boxes to use in the defense of liberty: soap, ballot, jury, ammo. Use in that order. Starting now.
  11. "4MB of DRAM" by vogon+jeltz · · Score: 3, Funny

    ... those were the times. Ahhh, memories!

  12. Only a PPC 440? by Anonymous Coward · · Score: 1, Interesting

    Why not more powerful CPUs? a 440 is hardly any kind of workhorse. A G4 at that speed would be too hot, but since PIII machines can run with just a small passive heatsink now wouldn't that have been a much better choice?

    1. Re:Only a PPC 440? by shaitand · · Score: 4, Informative

      No, It's a supercomputer. Those are RISC processors, a PPC 440 in reality gives better performance than a CISC processor like the PIII

    2. Re:Only a PPC 440? by Anonymous Coward · · Score: 1, Informative

      RISC vs. CISC is a moot point nowdays.

      A PIII is as much a RISC processor under the hood as the PPC, but neither are pure RISC.

      Pure RISC sucks, pure CISC sucks.

    3. Re:Only a PPC 440? by vicotnik · · Score: 2, Insightful

      Why did this get such a high score? Why not compare an old Sparc 4 against an Athlon 64 or a Pentium 4, the Sparc has a RISC processor so it should be faster, right? The PPC 440 might be faster for a number of reasons, and being RISC instead of CISC is hardly even among the most significant. That x86 has a crappy ISA doesn't mean CISC that has to be slower than RISC in general.

    4. Re:Only a PPC 440? by be-fan · · Score: 4, Insightful

      -1: Disinformative

      RISC vs CISC means very little these days. Most current CPUs have a core even more minimal than RISC chips, but present a CISC (in the case of x86) or RISC (in the case of the G5) interface to the outside. They used the PPC 440 for different reasons:

      1) IBM had to do significant custom engineering for it, and they own the PPC 440 core. That allowed them to use it to design an SoC.

      2) They needed to add FPU hardware, which is easier to do on a design they own. The PIII only has one FPU, while this chip as 2 FPUs. IBM had to add this to the design, because the regular PPC-440 has no FPUs.

      3) The PPC-440 was designed from the beginning to be an embedded CPU. At 1GHz, a stock PPC-440 consumes about 2.5W. Even a low voltage PIII consumes more than that.

      --
      A deep unwavering belief is a sure sign you're missing something...
    5. Re:Only a PPC 440? by sploxx · · Score: 1

      No, it's not THAT disinformative. Although RISC/CISC doesn't say everything about processors, there are a few things to consider:

      - RISC needs smaller die sizes. The free space can be used for more pipelining etc. => more speed!
      - Compiler's perform way better on RISC. If you do not code in assembler => more speed!

    6. Re:Only a PPC 440? by Anonymous Coward · · Score: 0

      I simple RISC assessment could have solved the issue. I think we should form a task force investigating human resource inefficiencies.

    7. Re:Only a PPC 440? by afidel · · Score: 1

      Cache efficiency goes WAY up with a CISC design and since cache size and memory latency are the number one bottlenecks in modern CPU designs CISC wins. Sorry but even the G5 is a lot more CISC like than a traditional RISC processor as far as ISA goes. At the core allmost all CPU's are a minimalistic RISC design as the grandparent poster stated.

      --
      There are 4 boxes to use in the defense of liberty: soap, ballot, jury, ammo. Use in that order. Starting now.
    8. Re:Only a PPC 440? by sploxx · · Score: 1

      IMHO, cache efficiency only goes way up if your code size shrinks by the usage of a more expressive instruction set. But it is not said that
      CISC instruction sets are more expressive than RISC ones, despite their name. Look for example (no highperf. computing, I know, I know, but anyway) at the ARMThumb. That's 16bit/instruction. Clearly RISC.

  13. 4 MB DRAM by Sensei_knight · · Score: 3, Funny

    Holly shit where do I buy on of thoes!

    1. Re:4 MB DRAM by otis+wildflower · · Score: 5, Interesting

      4MB per CPU, each with 2 processing cores, and an onboard memory controller.

      Final version to have 65536 CPUs.

      Smells like 256GB to me, which is pretty decent in _any_ book, especially if it lives on the same silicon as the CPU...

    2. Re:4 MB DRAM by shaitand · · Score: 1

      Yes by why dram? Has supercomputing stopped using faster static memory in favor of cheap but slow dynamic and cpu hog (those refreshs are a big part of why a p1 will run reasonably well side by side with a p4 when you stick a fast drive controller card in it for 90% of computing operations) as well?

    3. Re:4 MB DRAM by javiercero · · Score: 4, Informative

      Because they are used embedded DRAM, which although not as fast as SRAM it gives more storage in fewer transistors. This leads to a smaller die, and lower power/heat dissipation.

      If your p1 runs at the same speed than your P4 for 90% of operations, then there is something wrong with your computer! The HDD is not the bottleneck for most modern computers, as they have enough memory to minimize page faults for most common home computing tasks.... Startup times however may be equal since both machines have to get the data/program from HDD... once stuffs are in memory, buubbie P1....

    4. Re:4 MB DRAM by shaitand · · Score: 1

      Let me correct myself. When running WINDOWS both machines are around equal speed for 90% of operations. On windows the drive plays a much bigger part because regardless of how much memory you have the drive swaps. Also a P1 is using EDO Ram so it does not sap the cpu in refresh.

      Get two systems together. Both with the same amount of memory and the same windows OS. Then install an IDE controller in the P1 and put fast ide drives on both systems (a faster scsi controller will do as well). Put something to the tune of a pci based Geforced 4 in both. Boot them up and see how they perform on common tasks.

      You say te harddrive is not the bottleneck on most modern computers. I disagree, in a world where windows runs on 90+% of computers, and windows swaps regardless of how much memory you have, in fact swaps MORE if you have more ram, the hard drive IS the biggest bottleneck in the system. Furthermore I'd argue that without dynamic memory in your system eating clock cycles with x million refreshes a second, the processor is no longer the longer the biggest bottleneck in the system.

    5. Re:4 MB DRAM by vogon+jeltz · · Score: 1

      Well, "editors" and such ...
      On a more serious note, I probably was fast enough to strike an obvious joke (4MB of RAM, very funny).
      What I'd actually like to know:
      Wouldn't these babies, considering *4* FPUs per double core be *screaminggggggggg* on typical tasks like Fluent, Ansys, Abaqus, and in general any fpu-intensive task?
      Wouldn't these computers be a revolution (in the sense of the word) for companies looking for "the bang for the buck"?
      Just wondering...

    6. Re:4 MB DRAM by stevesliva · · Score: 1

      Why DRAM? eDRAM has a 4x density advantage over SRAM, so they're getting 4MB on-die instead of 1MB. You can read the paper by IBM's eDRAM developers for 15 pages of detail, if you wish. Mind your megabits versus megabytes... array designers always talk megabits. DRAM's not as slow as you think.

      --
      Who do you get to be an expert to tell you something's not obvious? The least insightful person you can find? -J Roberts
    7. Re:4 MB DRAM by shaitand · · Score: 1

      With DRAM the higher the clock the more wasted cpu clockcycles, it scales that way.

      Definately higher density at a lower price but I thought the idea of supercomputers was to build the most powerful machine possible at ANY expense?

      Not to say that just thinking of having even the prototype in my basement doesn't make me want to cream my jeans. But this is not news, if you want lots of memory real cheap you go with dynamic ram, if you want the fastest ram you go with static. Dynamic ram bears a double penalty however in that it has to be refreshed and actually refreshing it eats clock cycles.

  14. Subjective... by Decameron81 · · Score: 5, Funny
    "The prototype has 512 CPUs running at 700 MHz, and when finished the entire machine will have 65536 dual-core CPUs running at 1 GHz or more."


    Woah, this is the first time I think a box with 512 CPUs at 700 Mhz each one is crap.

    Diego Rey
    --
    diegoT
    1. Re:Subjective... by turgid · · Score: 1
      Woah, this is the first time I think a box with 512 CPUs at 700 Mhz each one is crap.

      Let me guess, you bought a Pentium 4? :-)

  15. One should read 512 MB DRAM by ezh · · Score: 1

    It's gonna be 512 MB for BlueGene/L(ite) and 1Gb for proper BlueGene

    1. Re:One should read 512 MB DRAM by ezh · · Score: 2, Informative

      It's gonna be 512 MB for BlueGene/L(ite) and 1Gb for proper BlueGene

      I mean, per node :-)

      AFAIK, 512 Mb is just too little for proper protein-folding calculations, while 1Gb provides enough capacity... And, of course, no swap is possible in this types of systems

  16. It's gorgeous... by mOoZik · · Score: 4, Funny

    I think I wet my pants.

    1. Re:It's gorgeous... by Anonymous Coward · · Score: 0

      Again?!?

    2. Re:It's gorgeous... by NoMoreNicksLeft · · Score: 1

      Not "wet", the proper phrase is "I think I just *creamed* my pants".

      Some day, I too will own a supercomputer, even if it is a 15 yr old Cray...

  17. Maybe we haven't heard of it. by Anonymous Coward · · Score: -1, Offtopic
    By now we've all heard about IBM's Blue Gene/L
    I haven't heard of it. Maybe you all shouldn't be so presumptuous. Considering that the average /.er describes himself as a skeptic, I'm quite surprised to see this kind of stuff being stated. Maybe only a small portion of us have even heard of it?

    If it's just a figure of speech, then I guess it's no big deal.
  18. But will it run Linux? by Anonymous Coward · · Score: 0

    Answer: sortof.

  19. What's new? by LeoDV · · Score: 4, Interesting

    So, you mean they're going to build a computer that's going to be bigger, faster and with higher number stats than the current #1? Shocking!

    Sorry about the sarcasm, I'm only asking to be proven wrong, but isn't Blue Gene just more of the same, only bigger? Big Mac was interesting because of how cheap it was and because it was the first of its kind to use Macs, the Earth Simulator was interesting because it brought back custom chips for supercomputing as opposed to off the shelf components, we've been reading about IBM's dishwasher-sized supercomputer, articles about efficient supercomputing, so what's new about Blue Gene, besides being newer and bigger?

    Once again I'm not bashing, I haven't read much of anything but the /. blurbs, so I'm asking, is it just a bigger supercomputer, or does it have any "real" innovations?

    1. Re:What's new? by javiercero · · Score: 5, Interesting

      Well... Earth Simulator was bigger and faster... as was big mac. Every advance in computer design and fabrication has been about "bigger and faster." It may sound trivial to you, but that is because you got no idea about what is involved about making things "bigger and faster."

      What is significant about blue gene is that is some sort of compromise between off the shelf parts (PPC based Processing elements vs. the Earth Simulator SX based custom vector PEs), and efficient interconnection (plain crappy cluster like the Big mac with a better interconnect at multiple layers starting with dual cores per die).

      In the end it all leads to the same goal: tackling bigger problems faster. So it may sound trivial but there is a lot of research going into this baby.

    2. Re:What's new? by tigertiger · · Score: 5, Interesting
      Beowulf-style cluster are a big waste in terms of additional circuitry and hardware that you do not need in a supercomputer, from I/O busses to the power supply. Of course, since off-the-shelf components are so cheap, it is still cheaper to buy the stuff than to design your own tailor-made circuitry - up to a certain scale.

      That is where IBM tries to go: BlueGene's design is based on a system-on-a-chip - everything (except memory) is integrated on a single chip. In the long run, this allows them to build systems much larger than you could with a Beowulf. They are basically aiming for a system where you can easily add computing power by simply putting in a few more chips, and the thing will scale. They are doing the same thing for storage with this brick

      BlueGene is a also the first supercomputer marketed to the life sciences. It's interesting to see that it developed from a project at Columbia University called QCDOC for "Quantum Chromodynamics on-a-chip" which did research in computational high-energy physics, and QCDSP before, which used DSP processors to build a supercomputer about ten years ago. Both an instructive example how academic research in the long run becomes industrially relevant, and how science changes.

    3. Re:What's new? by Jeff+DeMaagd · · Score: 1

      I thought the point of Blue Gene was to be faster and smaller, and that it follows the efficient computing idea where you get significantly more TF per watt, much more TF per cubic volume, much less cooling needed per TF, etc.

    4. Re:What's new? by stevesliva · · Score: 1
      That is where IBM tries to go: BlueGene's design is based on a system-on-a-chip - everything (except memory) is integrated on a single chip.
      Actually, even the 4MB of eDRAM is on the chip.
      --
      Who do you get to be an expert to tell you something's not obvious? The least insightful person you can find? -J Roberts
    5. Re:What's new? by Anonymous Coward · · Score: 0

      What is significant about blue gene is that is some sort of compromise between off the shelf parts (PPC based Processing elements vs. the Earth Simulator SX based custom vector PEs), and efficient interconnection (plain crappy cluster like the Big mac with a better interconnect at multiple layers starting with dual cores per die).


      Just to nitpick, a 440 is not an off-the-shelf product that you can 'just' put onto a board; you have to create/commission a chip that uses the 440 core. Also, the 440 does not have any FPUs; you have to add them to the custom chip you are building. Two make a dual-core chip, you have to put two 440 cores on a custom chip.

      Therefore, the chip that is used in BlueGene/L is a custom chip, with dual FPUs tacked-on, and is most certainly not off-the-shelf.

  20. Good story by Anonymous Coward · · Score: -1, Troll

    I quite enjoyed that, thanks. Better than the usual geek rubbish on Slashdot.

    Last night I masturbated thinking of my ex-girlfriend -- except that she had no arms and legs. I had removed them and was raping her. Her helpless sobbing mingled with my orgasm.

    Serves her right for cheating on me, I thought. But today I feel disgusted with myself. Is this wrong?

  21. Whoaa... by Anonymous Coward · · Score: -1, Redundant

    ... can you imagine a beowulf cluster of these?

  22. i'm scared to post ... by Anonymous Coward · · Score: -1, Redundant

    an 'imagine a beowulf cluster of these' post

  23. I'm surprised! by Epsillon · · Score: 4, Funny

    The standard of trolling has certainly fallen recently. Where's the SCO licence fee estimate for the finished 65536 processor SMP unit? You got a better class of idiot in those days... ;o)

    --
    Resistance is futile. Reactance buggers it up.
    1. Re:I'm surprised! by Ianoo · · Score: 4, Funny

      IBM better not forget to pay SCO their $45,809,664!!

  24. The Racks Are Not Angled by The+Other+White+Meat · · Score: 3, Insightful

    If you actually look at the picture, closely, you will see that the racks themselves are NOT angled to reduce active cooling.

    At the left side of the row of racks, there is an angled cover, which is either decorative, or being used to force cold air down the row of racks. Likely, its just decorative, and the cold air is being forced up from the raised flooring below.

    Just like it is in every other enterprise-grade computer room...

    --

    --- Generation X: The first generation to have SIG lines inferior to their parents... ---
  25. In Soviet Russia... by The+Other+White+Meat · · Score: -1, Redundant

    ... a beowolf would cluster YOU!

    --

    --- Generation X: The first generation to have SIG lines inferior to their parents... ---
  26. See for yourself by mangu · · Score: 4, Funny
    save this as "test.c":
    #include <stdio.h>
    #include <stdlib.h>
    #include <time.h>
    #include <sys/time.h>
    #include <unistd.h>

    #define N 1000

    struct timeval tv;
    struct timezone tz;
    long bu, bs, du, ds;

    int main()
    {
    int i;
    double t;

    printf("Infinite loop test\n");

    gettimeofday(&tv, &tz);
    bs = tv.tv_sec;
    bu = tv.tv_usec;

    for (i = 0; i > -1; i++)
    ;

    gettimeofday(&tv, &tz);
    du = tv.tv_usec - bu;
    ds = tv.tv_sec - bs;
    t = (double)ds + (double)du / 1e6;
    printf("executed in %13.6f seconds\n", t);
    }


    compile and link with:

    gcc -g -o test test.c

    run:

    Infinite loop test
    executed in 3.888419 seconds
    1. Re:See for yourself by Anonymous Coward · · Score: 0, Funny

      That's pretty funny, in a geek sort of way. Shit .. that means I'm a geek. Shit.

    2. Re:See for yourself by zora · · Score: 1


      [zora@dilaudid infiniteloop]$ gcc -g -o test test.c
      [zora@dilaudid infiniteloop]$ ./test
      Infinite loop test
      executed in 12.825751 seconds

      My computer sucks.......

      --
      In the end they will lay their freedom at our feet, and say to us, "Make us your slaves, but feed us." - Dostoevsky
    3. Re:See for yourself by uid8472 · · Score: 3, Funny

      My computer has 64-bit ints, you insensitive clod!

    4. Re:See for yourself by Anonymous Coward · · Score: 0

      Actually int by definition is only 32 bits, even on 64 bit machines. long's however are the word length of your machine, 32bit for i386, 64bit for alpha, etc.

    5. Re:See for yourself by Anonymous Coward · · Score: 0

      Absolute rubbish!

    6. Re:See for yourself by ajshankar · · Score: 0

      The only size required by the C spec is char: 1 byte. int can be anything you want, as long as it's >= short and = long.

    7. Re:See for yourself by The+Original+Atrox · · Score: 1

      Infinite loop test
      executed in 18.778423 seconds

      this sucks... Time to get a watercooler and overclock this baby...

      Microft

      --
      -Beware of he who would deny you access to information, for in his heart, he dreams himself your master.
    8. Re:See for yourself by Anonymous Coward · · Score: 0

      Bah my Sparcbook is greater than all of yours. It only took me 215.046200 seconds to complete. I'd get something faster if it ever actually broke and I didn't love it so.

  27. Dumpsterdiving seems a waste of time at IBM by Mentorix · · Score: 1

    There's a box in the background of this picture which has written on it "IBM Confidential trash".

    I guess corporate espionage is quite real for these guys.

    1. Re:Dumpsterdiving seems a waste of time at IBM by Anonymous Coward · · Score: 0

      yeh

    2. Re:Dumpsterdiving seems a waste of time at IBM by treat · · Score: 2, Funny

      In that picture you can also see sprinklers! Oh my.

    3. Re:Dumpsterdiving seems a waste of time at IBM by 888+Geek+Help · · Score: 1

      That guy?s a Romulan

      --
      -888 Geek Help (888-433-5435)
  28. No, *Smaller* and Faster by RalphBNumbers · · Score: 3, Insightful

    What's big about BlueGene/L is that it's small. That 512 processor prototype they mention in this article is the Dishwasher-sized computer you heard about.

    BlueGene/L is about driving down the cost of supercomputing, not only in terms of money spent on hardware, but in terms of space, cooling, and maintanance, while at the same time improving scalability.

    BlueGene/L is going to put 65,000+ processors in less space, using less power, and costing less, than many of todays >10,000 processor systems.
    They do this with a minimalist approach, each processor is a SoaC (System on a Chip), with everything from the memory controller to internode networking to two cores and 4FPUs on the die, and the only other thing in a node besides the processors is a bit of RAM. This allows them to use much less power per node and gives them less heat per node to dissipate, which lets them pack the nodes much closer, which cuts down on internode latency, which increases scalability.

    --
    "The worst tyrannies were the ones where a governance required its own logic on every embedded node." - Vernor Vinge
    1. Re:No, *Smaller* and Faster by RalphBNumbers · · Score: 1

      Dammit, my less than symbol should be pointing the other way.

      --
      "The worst tyrannies were the ones where a governance required its own logic on every embedded node." - Vernor Vinge
    2. Re:No, *Smaller* and Faster by dandiep · · Score: 1

      I don't even know if its so important that the computer is small. What is more important is what it allows us to simulate and what it is a step toward simulating. It allows us to simulate protein folding - which is a major step in developing drugs. It is also a major step toward simulating an entire cell or organelles of a cell. We need this computing power so we can see what effect a drug has without making it, running tests on humans/animals, etc.

      --
      Dan Diephouse
      http://envoisolutions.com/p
  29. Linux nucleus for slaved compute nodes? by Morgaine · · Score: 3, Interesting

    The part of the article that I found most interesting was:

    Linux actually resides on only a comparatively small number of processors; the bulk of the chips run a stripped-down operating system that lets it carry out the instructions of the Linux nodes.

    The "stripped down operating system" must be the distribution nucleus on the compute-only subnodes, presumably something that allow the Linux nodes to distribute the code and I/O of computations to them and to query or control their state during debugging, and to reaccquire lost processor control.

    It's only a matter of time before those of us who already have sizeable LANs at home will have embedded compute-only clusters within them too. Those would differ substantially from the typical Linux clustering for high availability. Instead of a non-Linux nucleus on those subnodes though, I'd prefer to see a pretty ordinary Linux kernel running slaved to remote masters.

    Is anyone already playing with something like this in their Linux clusters?

    --
    "The question of whether machines can think is no more interesting than [] whether submarines can swim" - Dijkstra
    1. Re:Linux nucleus for slaved compute nodes? by Anonymous Coward · · Score: 0

      Well, my openMosix cluster has a bare minimum linux install on all the nodes. I dare say that this makes it less efficient than some embedded setup, but as I've got such varied hardware on there (and no bootable network cards) it's the best solution I've come up with so far.

    2. Re:Linux nucleus for slaved compute nodes? by Anonymous Coward · · Score: 0

      Local hard disk drives create a bit of a management nightmare, and I've always wanted to avoid that. One live CD per box is reasonably management-free, so that was going to be my first step into proper clustering.

      Ideally though, I'd like to slot multiple vertical motherboards side by side for good heat flow, and use on-board LAN for all I/O, network boot implied. The main problem with this approach is the amount of bespoke construction involved.

      What's stopped me doing so though is that when you cost this up, it doesn't come out much cheaper than a commercial blade server, and it's a lot less flexible than they tend to be. So, at present, it looks like I'll grit my teeth and buy the real thing.

    3. Re:Linux nucleus for slaved compute nodes? by Anonymous Coward · · Score: 1, Interesting

      This is what BProc (The Beowulf distributed process space) is.

    4. Re:Linux nucleus for slaved compute nodes? by afidel · · Score: 1

      Most modern systems support PXE boot so you can lan boot even without a bootable NIC. Put together a minimal linux system with PCI probing and all the drivers needed for your various systems hardware and you should be able to go without local disks at all.

      --
      There are 4 boxes to use in the defense of liberty: soap, ballot, jury, ammo. Use in that order. Starting now.
  30. Can you imagine... by chrysalis · · Score: -1, Redundant

    a Beowulf cluster of those?

    (sorry, but someone had to say it)

    --
    {{.sig}}
    1. Re:Can you imagine... by HiThere · · Score: 1

      Can you take your own question seriously?

      This is a dense amount of processing power. Beowulf clusters aren't nearly this dense. Actually, if one were to create a "Beowulf cluster" of these, probably the Blue Genes would be attached to the Beowulf nodes, rather than being themself nodes. (I suspect that the controller for the Blue Gene is a bit specialized for controlling the Blue Gene.)

      Attaching a Beowulf Cluster of these together would result in a computer that was significantly slower...but able to deal with larger problems if there was some automatic way of partitioning those problems so that they could be parcelled out. If there isn't, then most of the increased power would end up wasted...except on a rather limited class of problems that were simple enough and regular enough that they could be pre-factored. Many of these problems are quite important (protein folding is probably one of them), but they are a small subset of "all problems".

      --

      I think we've pushed this "anyone can grow up to be president" thing too far.
  31. 24 CPUs per unit by chrysalis · · Score: 1

    Looking at the photographs, the entiere beast resides in 64 rack cases. With 42 units per case and 65536 CPU total, there are 24 CPUs per unit. Not bad :) I can't imagine the overall heat of the thing.

    --
    {{.sig}}
  32. great by arctan1701 · · Score: 0, Redundant

    wait till SCO hears about this...

  33. Marc Snir by thesilverbail · · Score: 1

    huh. the guy who initiated and led the Blue Gene project in the beginning was Marc Snir. But then I believe some major fallout happened and it underwent major technical and managerial changes. oh well.

    --
    I have found a truly wonderful proof of Fermat's Last Theorem, but unfortunately this sig is too small to contain it.
  34. Looks like thier going to keep this thing busy by Coolmoe · · Score: 0

    on a bunch of projects. Anybody get that list of projects that had to be split and alphabetized. Some of the projects seem a little silly and some seem to really have some really good uses to mankind. One of the things that made me laugh was to use it in computer virus research. I thought the internet was the biggest "viral research lab" around. How are you going to model that with a supercomputer? I may be spewing utter ignorance here however.

    --
    Got hosting
  35. Fucking troll moderators .... by Anonymous Coward · · Score: 0

    Bah.

  36. OT: Hire a photographer (me ;)! by MasTRE · · Score: 2, Insightful

    Besides the fact that their Nikon D100 has a stuck (hot) pixel, the pictures of people (first "set" on the page) are really bad quality-wise and there is not much creativity - i.e. two shots of the same geek (Hall) taking heatsink temperatures from slightly different angles aren't exciting even to fellow geeks.

    Other than that, keep up the great work IBM!

    --
    Must-not-watch TV!
    1. Re:OT: Hire a photographer (me ;)! by Dahamma · · Score: 1

      The first picture is my favorite... the classic "ok guys, point at the computer and pretend you're saying something important!" pose.

  37. You're all Missing the point here... by shrugwhaa · · Score: 2, Insightful

    I hear:

    Oh wow, another technical marvel

    Oh Gee, another super computer...

    Morons...

    The whole point here, is that it makes the simulation
    of folding a complete gene in about a years time.

    If THAT doesn't bowl you over, don't post.

    p.s. I can hear the rest of you "umm... so?" people and I can't help you. Sorry. :)

    1. Re:You're all Missing the point here... by PingPongBoy · · Score: 1

      The whole point here, is that it makes the simulation
      of folding a complete gene in about a years time.


      Isn't that folding a protein?

      But how big a protein?

      A chromosone would take a lot longer to simulate - but it's essentially a double helix when you get down to the molecular level. That would be a good test though - if you take DNA and can't simulate it going into a double helix how can you trust the computer?

      --
      Know your pads. One time pad: good for cryptography. Two timing pad: where to take your mistress.
  38. Wouldn't it be great... by mgeneral · · Score: 2, Interesting

    ...If they ported over VMWare to run on this bad boy? Imagine the number of guest OS's you could run. This thing could be the data center of all data centers.
    But otherwise, for all intents and purposes, its extremely proprietary and will ultimately run just a few specialized applications.
    Never-the-less, with virtualized computing and beheamoth systems like these, the future of data centers is sure to change.

    --

    Goals are deceptive - the unaimed arrow never misses.
  39. Cool movie... by gonzo_bozo · · Score: 1

    Maybe it's in part because I am currently studying these molecules but the movie is cool. It's got a nice cello background music and the resolution is well above average. I never thought watching a molecular simulation would make me feel that way.

  40. du ? I want to crack passwords!!! by edufortes · · Score: 1

    er... With this power i want to crack some MD5 's !!!!! :)

    --
    Eduardo N. Fortes
  41. Did they... by tesloni · · Score: 1

    ...decided to finally answer the great question of Life, The Universe and Everything?

    If they need help for that, they can read an Douglas Adam's "Hitchhikers Guide to the Galaxy".

  42. We NEED faster, dammit! by 23 · · Score: 1
    From where I come from, we do something called "lattice gauge theory", which basically simulates QCD, the force that e.g. binds quarks into protons and neutrons. This is an amazingly hard problem, which is not solvable or even tracktable with pencil and paper (theorists' favorite tool).


    Now, Ken Wilson (=cool Nobelprize guy) who basically started this field, gave a famous estimate in a talk at a conference some time back, stating that to do a serious project in lattice QCD, one would need some (listen up now...) PetaFlopYears. Yes, imagine a computer being able to 1 PetaFlop/s running for one whole year nonstop. Another guy, Karl Janssen, said last year at the same annual conference, "yup, still true". So there is the requirement. Now, where is the machine?


    From what I know, BlueGene is a direct decendant (sp?) of QCDSP (QCD on a DSP), a special purpose beast, built with IBM at Columbia Univ. to tackle, guess what, QCD. Its successor, soon to be rolled out, is QCDOC (= QCD on a chip), again together with IBM which is very close to the BG design. So all around cool IBM is using this know how and putting it to use for other research fields. If you're really interested, just google for QCDSP, QCDOC, lattice QCD, etc.


    I don't know anything about protein folding or weather, but I can imagine, those guys are just as anxious to get some serious CPU in form of these monsters as the lattice guys.


    As was pointed out above, "smaller" just means, you can cram more in one room and therefore have more CPU.


    PS: not only do I think IBM is geeky-cool, I'm starting to admire their strategy: do worthwhile research with Academia, use know how, make into money, cf. Grid aka. IT as utility, BlueGene aka. Supercomputer in a box. awesome.

    1. Re:We NEED faster, dammit! by tigertiger · · Score: 1
      I'm starting to admire their strategy
      I concur - I used to be in lattice gauge theory and moved into computational biology :-). You have to follow the money, and IBM is very good at this. Also, biology is more fun, they are just starting to solve problems seriously on a computer.