supercomputer
Latest
Shall we play a game?
When you think of a rapidly developing Asian nation, fast becoming a global superpower and now harnessing the potential of supercomputing, what's the first purpose for this supercomputer that comes to mind? Hint: think video games, not war games. Regardless of your first thought, we're betting you wouldn't have guessed "World of Warcraft"... but it's true. China's World of Warcraft servers are run by The9, who own Blizzard's distribution rights in the country. "Earlier this year, The9 boasted of hosting more than one million World of Warcraft players online at the same time.To support the complex calculations required to create the game's graphics, The9 owns more than 10 supercomputer systems," Ashlee Vance reports for The New York Times.
NVIDIA announces cost, energy-saving Tesla Personal Supercomputer
AMD has already outlined its plans to harness the power of its GPUs for some added computing muscle, and it looks like NVIDIA is now taking things one step further by announcing its new GPU-based Tesla Personal Supercomputer, which promises to deliver the power of a traditional supercomputer cluster at 1/100th of the price. That "personal supercomputer" is actually a platform based on NVIDIA's new Tesla C1060 GPU Computing Processor, which itself is based on NVIDIA's CUDA parallel computing architecture. The supercomputers themselves will come from a whole host of manufacturers that have already partnered with NVIDIA, including ASUS, Dell, Lenovo, and a number of more specialized computer makers. While complete details on those systems are still a bit light at the moment, they'll apparently be "priced like a conventional PC workstation," and the first few out of the gate should be available starting today.
Roadrunner beats Jaguar in TOP500 supercomputer rankings, cartoon antics strangely absent
While titles like "world's fastest" and "world's largest" are fleeting at best, it's rare that we see such things taken down this quickly. Last week Cray delivered a big dish of braggadocio, talking up its 1.64 petaflop XT Jaguar supercomputer as the fastest (non-classified) machine in the world. Now, like some rocket skate-wearing coyote who's run out of thrust, it's been stymied by IBM's Roadrunner, deployed at Los Alamos. TOP500 is the authority on these matters and that list's latest rankings place Roadrunner in first place with a speed of 1.105 petaflops; Cray's Jag comes in second with a paltry 1.059. What about that 1.64 figure from last week? That was the hypothetical limit, and while it did deliver real-world performance of 1.3 petaflops for the folks at Oak Ridge, TOP500 relies on the Linpack benchmark for its ratings and apparently the Jag just couldn't deliver the goods there. Perhaps, Cray, it's time to make another call to ACME -- or AMD as it were.
Cray supercomputer is world's fastest (that we're allowed to know about)
Seattle-based Cray has been wowing us with massive, ridiculously fast supercomputers since the '70s, establishing a position for its machines high on every geek's most wanted list -- despite never publishing a Doom benchmark. In recent years the title of "world's fastest" supercomputer had been lost by the company, ping-ponging from Wako, Japan to Armonk, NY, but is now back in Cray's hands with the implementation of the XT Jaguar. It's comprised of over 45,000 quad-core Opteron processors, 362GBTB of memory, and has a 10PB (petabyte) storage array, able to perform calculations at a massive 1.64 petaflops. That's over one and a half quadrillion operations per second and more than 50 percent faster than IBM's previous heavyweight. Mind you, Cray is quick to point out that this is the fastest machine being used for non-classified research, a caveat that just makes us even more curious about what's keeping the Pentagon's server rooms warm and loud... and apparently orange. Update: That should be 362 terabytes of memory, not mere gigabytes.[Via UPI.com]
World's largest computing grid lives to go live
Contrary to popular belief, the world as we know it didn't implode after the Large Hadron Collider was flipped on. Sure -- someone, somewhere is growing a ninth arm and trying desperately to land a cameo on Fringe, but the planet at large is still humming along just fine. Now, the world's most ginormous computing grid (the Worldwide LHC Computing Grid, or WLCG) has gone live, and the gurus behind it are celebrating the beginning of its momentous data challenge: to analyze and manage over 15 million gigabytes of data each year. The Grid combines the IT power of over 140 computer centers, 100,000 processors and the collaborative efforts of 33 countries. Unfortunately, there's no word on when the official WLCG-based Call of Duty 4 server will be green-lit for action, but we hear it's pretty high on the priorities list.[Via China View]
Microsoft and Cray deliver "mainstream" CX1 supercomputer: starts at $25k
C'mon, who here doesn't want their very own supercomputer to do, um, whatever they want with? In an effort to make sure every man, woman and child has an absurdly powerful number cruncher in their home (let's go with OSPP, or One Supercomputer Per Person), Microsoft has tag-teamed with the fabled Cray in order to "drive high productivity computing into the mainstream." The Cray CX1 Supercomputer comes loaded with Windows HPC Server 2008 and incorporates up to 8 nodes and 16 Intel Xeon CPUs (dual- or quad-core); additionally, it boasts up to 4TB of internal storage, 64GB of memory per node and interoperates nicely with Linux. The CX1 is said to be the most affordable supercomputer offered by Cray (not to mention the "world's highest-performing computer that uses standard office power"), but it'll still run you anywhere between $25,000 to well over $60,000. Chump change, right?[Via NetworkWorld]
Supercomputer Huygens beats Go professional: no one is safe
You know how Go nerds are always going on about how magical they are since supercomputer AI hasn't yet cracked the ancient board game, and rarely beats even an average Go player? No? Maybe those are just our nerdy friends. Well, those folks can wipe the smug grins off their faces as they're faced with the sobering reality of defeat: Dutch supercomputer "Huygens" has defeated a human Go professional in an official match at the 24th Annual Congress of the game Go in Portland, Oregon. The newly-minted supercomputer was aided by the recently-developed Monte-Carlo Tree Search algorithm, a whopping 60 teraflops of processing power and a considerable 9 stone handicap. Poor Kim MyungWan -- who managed to beat the computer in three "blitz" games leading up to the actual match, and probably won't be hanging up his Go hat just yet -- didn't stand a chance.[Via Tech Digest]
WoW Chinese distributor The9 dominates China's supercomputers
One thing that your average WoW player probably doesn't think about much is the sheer processing power that's needed to play the game. I don't mean your gaming rig personally. WoW's actually pretty forgiving on that front. I mean the server hardware over on Blizzard's end, the stuff that makes us miss some gaming every other Tuesday or so. Do we really stop to think about how powerful it is? It turns out that it's powerful enough to dominate a list of China's top 100 supercomputers. The Register reports that 5 of the top 10 supercomputers in China are owned by The9, WoW's China distributor. Not only that, it also owns at least 12 of the top 100 overall, and perhaps more. That means that more than 10% of China's best supercomputing power is directed toward MMORPGs such as World of Warcraft.
PS3's Cell processor helps locate oil
Besides the Roadrunner and our beloved PS3, the Cell processor can now add another task it can do into its repertoire. Repsol, one of the world's ten largest oil companies, is employing the use the of the Cell processor, more specifically the PowerXCell 8i variant on a BladeCenter QS22 computer, to "analyze undersea rock formations in the search for untapped reserves." An analyst commented that 75% of the reserves are in deep water, which are 30,000 feet below and are estimated to hold 56 billion barrels of oil. This amount would meet the entire U.S. demand for oil and gas for about five years. The use of a technique called reverse time migration will allow Repsol to survey regions as much as six times faster and reach areas that were inaccessible before.What do you guys think about this? Will this lower gas prices somewhere far, far away from $5 a gallon? Are you going to mod your PS3 to start drilling for oil in your backyard? Ah, the possibilities...
Fastest-ever Windows HPC cluster nets 68.5 teraflops
In the grand scheme of supercomputers, a homegrown cluster constructed by Microsoft is just a boy among men. However, said rig has set at least one record by becoming the "fastest-ever Windows HPC cluster." At the International Supercomputing conference in Dresden, Germany, it was announced that this beast ranked 23rd in the world "with a problem-solving performance of 68.5 teraflops." The National Center for Supercomputing Applications utilized a beta version of Windows HPC Server 2008 to hit the aforesaid mark, and if you're curious as to what it took to get there, try 9,472 cores of processing power. There's more where this came from for the hardcore nerds in attendance right in the read link.[Via Slashdot]
IBM creates new supercomputer with Cell processor
The Cell processor, known as the heart of the PS3, is being used every day in rather extraordinary situations. IBM has crafted yet another supercomputer, codenamed Roadrunner, which runs at a record-setting petaflop speed. (A petaflop is the equivalent of one thousand trillion calculations per second.) This speed makes Roadrunner twice as fast as the previous best supercomputer."We are getting closer to simulating the real world," Bijan Davari, vice president of next generation computing systems at IBM, told BBC News. The computer will be installed in a US government laboratory to monitor the US nuclear stockpile. It will also be used for research into astronomy, genomics and climate change.The Roadrunner will use less than 20,000 chips, and will use a hybrid design of both conventional supercomputer chips, and PS3's Cell processor. The next design by IBM will be nicknamed "Skynet."[Thanks, Richard!]
World's fastest: IBM's Roadrunner supercomputer breaks petaflop barrier using Cell and Opteron processors
When you're looking to set a record this is how you do it. Not only has IBM's Roadrunner supercomputer come on-line, it's now the world's fastest -- twice as fast as the old BluGene/L champ -- and churning through 1.026 quadrillion calculations per second. The $133 million supercomputer achieved the milestone with the help of 12,960 "improved" Cell processors (yes, like those powering your PS3) and a smaller number of AMD Opteron processors -- 116,640 processor cores in total. Unfortunately (or fortunately depending upon your perspective), Roadrunner is for military use only so you'll have to solve the traveling salesman problem on your own time. While not quite into Exaflop territory, we're definitely on the way.[Thanks, Chris S.]
Researchers create supercomputer with four GeForce 9800 GX2 cards
It's far from the first supercomputer created with the help of some gaming hardware, but this rig built by a group of researchers from the University of Antwerp is certainly impressive enough in its own right, with it employing four of NVIDIA's high-end GeForce 9800 GX2 graphics cards (which combined pack eight GPUs) to help develop new computational methods for tomography. Dubbed the FASTRA, the system also packs an AMD Phenom 9850 processor, 8GB of RAM, and 750GB hard drive, all of which is powered by a 1,500W power supply (and tastefully lit up with some blue LEDs). That apparently lets 'em do calculations that previously took an hour in just a few seconds, not to mention finally get a decent frame rate in Crysis. Be sure to check out the video after the break for a thorough (and more entertaining than it should be) overview of the system.
Researchers tout 20 million processor-strong supercomputer to study climate change
It looks like a group of researchers at UC Berkeley have come up with a rather unique way of solving the problem of getting supercomputers past the processing power / energy consumption barrier, with them now touting the possibility of using millions of low-power embedded microprocessors instead of conventional server processors. That tantalizing prospect has apparently already lead to a deal with Tensilica Inc, which will provide the Berkeley researchers with some of its Xtensa LX extensible processor cores to use as the "basic building blocks in a massively parallel system design." Ultimately, the researchers say they could one day build a massive supercomputer consisting of 20 million embedded microprocessors at a cost of $75 million, which they say would have a power consumption of less than 4 megawatts and a peak performance of 200 petaflops. That, they say, would be enough for it to create climate models at 1-kilometer scale or, as the researchers put it, more than 1,000 times more powerful than anything available today.[Via TG Daily]
IBM plots global-scale shared computer to host entire internet as application
Well, it sure looks like IBM is keeping all its supercomputing bases covered these days, with it not only working on a chip-sized supercomputer, but a global-scale shared computer that'd be capable of "hosting the entire internet as an application." That latter word comes in the form of a white paper ambitiously dubbed "Project Kittyhawk" (we're guessing they found "Project Mulitvac" a little too obvious), which aims to explore the construction and implications of such a massive scale computer. That apparently wouldn't be a SETI or Folding@home-style shared computer consisting of everyday PCs, however, with it instead relying on IBM's petaflopping Blue Gene/P as a common platform, which would effectively be able to run any web-scale application you could throw at it. Of course, none of that has moved very far beyond the page just yet, so you can rest easy that there's still no supercomputer out there that's capable of bringing the entire internet to the halt on a whim, at least that we know of.[Via Slashdot]
The PS3 takes the cape of a supercomputer once more
Like Clark Kent, it's time again for the PS3 to discard its nerd-goggles and rip that button-down shirt off to reveal what lies beneath: a superhero. Or a supercomputer, if you felt the comparison went a bit far. A professor at the University of Massachusetts at Dartmouth, Guarav Khanna, bought eight PS3's and clustered them together. Using Linux, he and his colleagues are enjoying some heavy-duty scientific research thanks to their newfound supercomputer.Khanna offers up the notion the supercomputer could be used outside of scientific research. See, most supercomputers have intense calculations queued up and answers can take a day or two to get back to researchers. If every school had a PS3, or this chain-gang of PS3's as it were, that problem would go away. It could also double as a computer running Linux allowing kids to do work and surf the web. Obviously. Now, PS3, go get your glasses and shirt back on, because we want to play some Folklore.[via PlayStation Forums]
Britain's new fastest supercomputer makes its ominous debut
It's always tough to keep track of the fastest supercomputers around at any given moment, but according to Reuters, the rather imposing block of machines pictured above is now at the top of the heap in Britain, and at least near the top in all of Europe. Dubbed HECToR (for High-End Computing Terascale Resource), the Cray-built rig packs the equivalent power of 12,000 desktop PCs, or enough to perform some 63 million calculations per second, which'll apparently be used to study everything from climate change to new medicines -- or, as the researchers responsible for the computer put it, "push forward the boundaries of knowledge." In related news, Engadget's Folding@home team is still accepting new members -- nudge nudge, British friends.[Thanks, Matt]
IBM creates a chip-sized supercomputer
Good news, everybody! Those super-geniuses over at IBM have whipped up a new form of CPU transfer which utilizes pulses of light instead of electricity to move data between cores on a chip. The new technology -- which is one-hundred times faster than current speeds -- is called silicon nanophotonics, and if implemented, could downsize supercomputers to laptop stature. The invention is unhindered by common problems with electrical chips, such as overheating and breakdown of data on short trips, allowing signals to pass unmolested over greater distances. Using this process, data can be moved a few centimeters, while requiring one-tenth as much power, resulting in lower operational costs for supercomputers. Will Green, a researcher at IBM, says that the company's creation will, "Be able to have hundreds or thousands of cores on a chip," and will result in huge speed boosts. Unfortunately, the project is on track to be carried out in 10 to 12 years, which leaves a lot of time to ponder if the chips will play Doom.
SiCortex intros SC072 Catapult -- 72 processor cluster for $15000
Off hand, we can't think of how we'd truly utilize the horsepower generated by a 72-processor cluster shoved into a "whisper-quiet, low-power deskside cabinet," but we'd happily draw up a plan if forced. SiCortex -- the same folks who delivered the bicycle-powered supercomputer -- has introduced its new SO072 Catapult, which features a standard Linux environment, 48GB of RAM and a trio of (optional) PCIExpress slots. This aptly categorized high performance computer (HPC) sucks down less than 200-watts of power, sports a pair of gigabit Ethernet ports and has room for six internal hard drives. Reportedly, each of the 12 SC072 nodes is a multi-core chip with six CPU cores, and while $15,000 may seem steep for your average tower, we'd say this is a pretty good value considering the hardware.[Via Gadgetopia]
IBM's BlueGene/L: world's fastest supercomputer, 3 years running
The TOP500 supercomputing list was just announced and IBM's BlueGene/L system has kept its crown. In fact, IBM's and the Department of Energy's co-developed monster at Lawrence Livermore has occupied the number 1 position since 2004. Of course, an upgrade was required boost the Linpack benchmark to 478.2 TFlop/s from the 280.6TFlop/s the machine was clocking just 6 months ago. The top 10 swath is dominated by the US, Sweden, and Germany with India breaking into the list for the first time at the number 4 position with its HP Cluster Platform 3000 BL460c system measuring 102.8TFlop/s. [Via Impress]
