supercomputer
Latest
NEC touts "world's fastest vector supercomputer"
NEC looks to have earned itself some new bragging rights in the supercomputer club with its new SX-9 model, which it claims is the "world's fastest vector supercomputer" on the market today. Helping it earn that distinction is a peak processing performance of 839 teraflops, and a peak vector performance of more than 100 gigaflops per single core (apparently a first for any supercomputer). That, NEC hopes, should make the SX-9 ideal for a wide range of uses, including weather forecasting, aerospace, the environment and fluid dynamics. No word on what it'll cost, but those looking to check out all that teraflopping for themselves should head to the Supercomputing 2007 expo in Reno, Nevada next month, where the SX-9 will make its public debut.[Via Physorg]
Supercomputer systems replaced by PS3
Cell's immense power is largely undisputed by the scientific community. Folding@Home is just one example of how Cell and the PS3 can compute massive amounts of data for the sake of scientific research. An astrophysicist recently linked eight PS3 systems together and was able to compute the same amount of data as 200 supercomputers. After Gaurav Khanna created code optimized for the PS3, Sony took attention and donated a few systems to his research."Basically, it's almost like a replacement," he told Wired. "I don't have to use that supercomputer anymore, which is a good thing." While one PS3 costs $500, a single supercomputer can cost $5000. The cost of eight PS3s is incredibly negligible compared to the massive cost of traditional supercomputers.With his PS3s at hand, Khanna will be researching gravitational waves.[Via Digg]
UMBC, IBM team up to create Cell-based supercomputer
We've certainly seen PlayStation 3s used in academia before, but a "significant gift" by IBM to the University of Maryland-Baltimore County is enabling the institution to further develop its Multicore Computational Center. The facility now utilizes 32 Cell processors, and researchers at the lab are getting set to "develop applications" to harness all that power. Reportedly, the supercomputer will dabble in "visualization and computer modeling applications for aerospace / defense, financial services, healthcare / life sciences industries and weather." Hit the link below for a video report.[Thanks, Ryan O.]Read - IBM, UMBC partnership Read - Video report
The bicycle-powered supercomputer
In an attempt to illustrate the low power requirements of its latest Linux-based supercomputer, a company called SiCortex put together a demonstration showing the massive number-cruncher performing billions of calculations per-second while a team of between eight to ten bicyclists pedaled to keep it going. The NextFest presentation was a ploy to showcase the small power consumption of the new machine, and to point out that energy conservation for both system power and cooling are major issues facing manufacturers and end-users. As the custom Trek bikes churned out 260-watts of power each, Dr. John Mucci -- the company's CEO -- said, "Ten years ago, this analysis was impossible on even the biggest computers, and now it's being done on a bicycle powered machine," then added, "Ten years from now, we'll probably be doing it right down at the docks as they unload the ships." Which sounds kind of weird, but isn't.[Via I4U]
Calvin College duo creates cheap, portable supercomputer
Just months after scientists were able to run a quantum computer simulation on an everyday PC, we're now hearing that a Calvin College student / professor tandem have created an inexpensive, portable supercomputer for crunching massive chunks of data on the go (and on the cheap). Dubbed Microwulf, the wee beast is hailed as a "machine that is among the smallest and least expensive supercomputers on the planet," and when not being checked as baggage on a flight, can reportedly process 26.25 gigaflops of data per second. The system itself touts "four dual-core motherboards connected by an eight-port gigabit Ethernet switch," and when initially constructed, it cost just $2,470 to build. Talk about a solid price-to-performance ratio.[Via Slashdot]
Scientists perform quantum computer simulation on vanilla PC
We've seen what (little) a quantum computer can do, but a pair of curious scientists flipped the equation around and sent a humdrum PC to do a supercomputer's work. Professor Peter Drummond and Dr. Piotr Deuar were able to "successfully simulate a collision of two laser beams from an atom laser using an everyday desktop computer," which would typically only be attempted on a substantially more powerful machine. Notably, the achievement wasn't entirely without flaw, as the purported randomness in the testing eventually "swamped everything" and forced the simulation to be halted in order to gather any useful data whatsoever. Unfortunately, we're all left to wonder exactly what kind of machine was used to chew through such grueling calculations (Compubeaver, perhaps?), but feel free to throw out your suggestions below. [Via Physorg, image courtesy of ACQAO]
IBM unveils one-petaflop Blue Gene/P supercomputer
It was but two years ago that we all sat in awe of IBM's beautiful 280.6-teraflop masterpiece, and it's finally safe to call that workhorse a thing of the past. Today, Big Blue announced the second-generation of its famed supercomputer (dubbed Blue Gene/P), which reportedly comes close to "tripling the performance of its predecessor." In an attempt to showcase its might, IBM suggested that 27 million clinical trials could be computed "in just one afternoon using a sliver of its full power." Additionally, the engineers went the extra mile by crafting a highly scalable system that comes in "at least seven times more energy efficient than any other supercomputer," but we're sure those 294,912 processors (each touting four PowerPC 450 CPUs) can churn through some energy when utilizing the whole petaflop. And just think, Sun probably thought the 500 or so teraflop production from its forthcoming Ranger would be the (momentary) top dog.Read - IBM's Blue Gene/P, via CNETRead - Sun's Ranger, via Wired
Researchers tout breakthrough in single chip parallel processing
Researchers at the University of Maryland's A. James Clark School of Engineering have developed a prototype of what they say could be the "next generation" of personal computers, one that's apparently 100 times faster than current desktop PCs. That considerable feat was made possible though the use of parallel processing on a single chip, in this case, cramming 64 processors onto a circuit board the size of a license plate. Just as importantly, the researchers also developed the necessary software to ensure all that computing muscle gets along, which they say makes the system "feasible for general-purpose computing tasks" for the first time. They don't appear to be content with things just yet though, saying that the same principles could one day be applied to systems with 1,000 processors on a chip the size of a finger nail.
IBM's BlueGene L supercomputer simulates half a mouse brain
Efforts to model the human brain (on IBM's Blue Gene, ironically) haven't reached the point of finality just yet, but it looks like the supercomputer has already tackled a smaller, albeit similar task at the University of Nevada. The research team, which collaborated with gurus from the IBM Almaden Research Lab, have ran a "cortical simulator that was as big and as complex as half of a mouse's brain on the BlueGene L," and considering that it took about 8,000 neurons and 6,3000 synapses into consideration without totally crashing, it remains a fairly impressive achievement. Notably, the process was so intensive that it was only ran for ten seconds at a speed "ten times slower than real-time," and while the team is already looking forward to speeding things up and taking the whole mind into account, it was noted that the simulation (expectedly) "lacked some structures seen in an actual brain." Now, if only these guys could figure out how to mimic the brain and offer up external storage to aid our failing memories.[Thanks, Richard L.]
Xilinx, Altera showing off FPGA coprocessors at IDF
While Field Programmable Gate Arrays (FPGA) haven't benefited from a good deal of buzz just yet, things could be taking a turn, as both Xilinx and Altera caught a few eyes at IDF. We've already seen the unique, albeit highly specific chips in a supercomputer, but the dedicated coprocessor / accelerator modules could be landing beside your Intel Xeon CPU. Essentially, the devices plug "directly into the processor socket of dual- or quad-socket servers" in order to provide "high performance application acceleration ranging from 10x to 100x compared to processors alone, while simultaneously reducing overall system power consumption." The modules act as targeted CPUs, effectively computing very specific tasks in a much more efficient fashion than a general microprocessor can alone, which could boost the speed of scientific, financial, and life science applications that rely on very particular calculations. Of course, mainstream adoption still has quite a ways to go, but the quicker we get dedicated physics and AI coprocessors to go along with these snazzy new GPUs, the happier (and poorer) we'll be.[Via RobotSkirts]Read - Altera demonstrates FPGA at IDFRead - Xilinx demonstrates FPGA at IDF
IBM pushing vertical stacking in next wave of supercomputers
Next-generation cooling technology isn't the only thing IBM's R&D crew is spending time with, as the chip giant has recently made plans to hit up "vertical stacking technology" in order to make the next wave of supercomputers really crank. Supposedly, "laying chips vertically -- as opposed to side by side -- reduces the distance data has to travel by 1,000 times, making the chips faster and more efficient." The new format will place chips directly atop one another and connect them with "tungsten filled pipes etched through the silicon," which will subsequently eliminate the need for wires and increase the speed at which data can flow. The questionably-dubbed "3D chips" will reportedly operate around 40-percent more efficiently than existing renditions, and considering that Intel is purportedly cooking up a similar agenda in their own labs, that "end of 2007" release date is quite likely to be accurate.
Tyan's 40 CPU Core Personal Supercomputer -- now shipping
Sure, you probably don't currently have a need for 256 gigaflops of processing power in your home. Rest assured, you will -- just as certain as the day when robots will care for our children and decorate our Christmas trees. So go 'head, raid the education fund and future proof your domicile by picking up a TYANPSC T-650 series computer from Tyan -- "PSC," as in Personal SuperComputer. Prices start at just $20,000 for a 40 CPU Core (10x quad-core Intel Xeons) system configurable with up to 60GB of RAM and a power draw of just 15A. And it's "portable" in the same way a refrigerator is portable: it has wheels. Who knows, maybe you can put those spare cycles to use by solving the traveling salesman problem. Better yet, join Engadget's Folding@Home team for some hardcore protein folding fun.[Via 64-bit-computers]
University of Edinburgh crafts energy efficient FPGA supercomputer
Considering that ATI and NVIDIA don't seem to be making any substantial strides in reducing the amount of energy required to run their products, it's a tad surprising to hear of an entire supercomputer running a bit leaner than the competition. Hoping on the ever-growing green bandwagon, University of Edinburgh developers are at it again, this time crafting an uber-speedy machine that's reportedly "ten times more energy efficient and up to 300 times faster than its traditional equivalents." Based on field programmable gate arrays (FPGA), the chips are not only very difficult to program, but they can currently only be used "to perform very specific tasks." Of course, the creators are more interested in the extreme number crunching and power saving abilities than anything else, and while no commercial uses have been identified just yet, the machines could purportedly be used in fields such as "drug design, defense and seismology."[Via CNET]
NC State engineer crafts academic cluster with PlayStation 3s
While universities have been cranking out supercomputers and research clusters for some time, an associate professor at NC State is utilizing IBM's highly-touted Cell processor in a slightly different form to craft his own farm. Similar to the Xbox Linux cluster from years past, this concoction consists of eight PlayStation 3 consoles networked together and powered via Linux in order to handle ridiculous amounts of number crunching. Dubbed the "world's first" PS3-based academic cluster, the creation boasts the ability to utilize "64 logical processors," and is set to be used to handle various research tasks when sly CSC students aren't firing up a round of Ridge Racer 7 after hours. Nevertheless, Dr. Frank Mueller noted that the biggest limitation in its current state is the "512MB RAM constraint," but did insinuate that he might try retrofitting additional memory if future tasks deemed it necessary. Still, we can't help but wonder how many spots the Pack could jump in the RIAA's Most Wanted list if this thing became a dedicated torrent server.[Via TWW]
BMW getting supercomputer; iDrive still sucks, though
BMW, one of the world's top auto makers, has decided that it needs an equivalent amount of horsepower for its supercomputers in its Switzerland development center, and thus struck a deal with Intel to purchase the Albert 2, a new supercomputer that works at 12.8 teraflops per second to help crunch numbers for the next generation of BMW engines and cars. That's not the fastest in the world, but it recently ranked 60th on the Top 500 supercomputer list -- a pretty respectable showing. But BMW and Intel are thinking bigger than just one supercomputer, as Intel will also get a huge sponsorship for BMW's Formula 1 cars (pictured), and in exchange the chipmaker will switch up every computer under BMW's roof to Intel-powered machines. Regardless of how fast the Albert 2 is, BMW is going to need to switch into fifth gear if it's really going to present the first Albert 2-designed F1 car, as planned, by January 16.P.S. -Some of us at Engadget do actually love iDrive. We're just playing around, BMW.[Thanks Terry B, via Heise]
University of Florida scientists build a faster supercomputer for spacecraft
Haven't you ever sat in your space shuttle/module/station and said to yourself, "Gee, I wish that we had faster computers like those terrestrial scientists do." No? Well our actual astronauts apparently have. See, while you've got your current dual-core (soon to be quad- or oct-core) desktop PC, computers in space have to endure a great deal more stress -- you know, like that whole launching into space thing, not to mention cosmic radiation, and a whole host of other rugged requirements, which takes a toll on what processors can be used. Engineers at the University of Florida (including Alan George, a professor of electrical and computer engineering, pictured at right) and Honeywell Aerospace announced late this week that a new supercomputer 100 times faster than any current space-bound computer (that's 20 processors at a combined power of 100 gigaflops) is under development. If all goes according to plan, it'll get hitched to an unmanned NASA rocket aboard a test mission in 2009.[Via Roland Piquepaille]
Stanford University tailors Folding@home to GPUs
Apparently the insane amount of gigaflops that your modern-day graphics card can churn out is nothing short of a phenomenon, as Folding@home's forefather Vijay Pande has tailored a new piece of software to harness to raw processing power of GPUs. Pande claimed that even the latest dual-core CPUs can't hold a candle to the floating point performance of ATi's X1900 / X1950 graphics cards. He estimated a Core 2 Duo chip could push about 25 gigaflops of folding power, while a high-end off-the-shelf ATi card could unleash a whopping 375 GFLOPS, which is about "20 to 40 times more speed" than the project has seen thus far. The team has also optimized the algorithms in the GPU-centric software, which is expected to add "10 to 15 times" more speed on top of the GPU's already impressive performance figures. Currently, the beta version is limited to the X1900 lineup, but plans are to include the X1800 variety in the near future, and Pande even mentioned that a PlayStation 3-friendly version was in the works. So if you aren't too busy tweaking your GPU-based supercomputer (or stressing over your energy bill), why not put those excessive GFLOPS to good use through Engadget's own Folding@home team, yeah?
Peakstream software taps into GPU for supercomputing power
While building a supercomputer has been whittled down to a science, Peakstream has developed a suite of applications that look towards those speedy PCI Express slots -- not the CPU socket -- for an extra boost of power. The company boldly states that a supercomputer can be created by harnessing the power of "common CPUs combined with the resources of modern graphics cards" to increase performance by "20x." This extreme form of load balancing exploits the tremendous potential housed in today's GPUs in order to schedule workloads, offload tasks onto the optimal processor(s), and manage calculations to minimize the queue of tasks to be completed. Granted, the biggest boon of a graphics processor is the extraordinary floating-point performance; for instance, ATi's X1950 XTX pumps out 750 GFLOPS in dual-graphics mode, while it'd take 31 Intel Xeon 5100 CPUs to crank out those same figures -- thus Peakstream feels that mathematical and computational applications (sorry, Doom fans) are best suited for its software. While having your own personal supercomputer churning those Engadget Folding@home cycles would be mighty impressive, the average joe isn't apt to drop $2,000 (per node) for Peakstream's suite, but maybe this explains the real intentions behind those 200 watt, energy sucking, externally-housed graphics cards after all.
IBM building world's fastest supercomputer using Opteron and Cell processors
One reason there's so much fuss (and delay) over the upcoming PlayStation 3 platform is the fact that it sports those new Cell processors jointly developed with Sony and Toshiba. Now those Cell procs are about to find themselves pumping away at the heart of a new $35 million supercomputer for Los Alamos National Laboratory. But this won't be just supercomputer mind you, IBM is hoping to reclaim the title of the world's fastest once completed in 2007. Dubbed Roadrunner, ACME IBM plans to jump from 280-teraflops to a full petaflop performance by combining AMD Opteron blade servers and Cell-based accelerator systems. A performance threshold achieved earlier this year by NEC's MDGrape-3, the supercomputer behind new pharmaceutical drugs and the curious taste and powerful punch of Mad Dog's 20/20 Red Grape Malt beverage. Mmm, grape.
Spinnaker project mimics human brain
Scientists have been on the hunt for years attempting to model the human brain, but typically any research in this area was strictly used for medical purposes. Steve Furber of the University of Manchester is the head honcho of the Spinnaker (short for spiking neural network architecture) project, which seeks to better understand the complex interactions of brain cells in hopes of creating more "fault-tolerant computers." Furber states that although we lose one neuron per second during our adult life (just think, you've already lost a couple dozen reading this post), yet our brains don't suffer any catastrophic meltdowns or periods of underperformance because of it. Rather, the brain has an impeccable ability to disregard malfunctions and find alternate ways to function without throwing a theoretical BSOD. The Spinnaker computer consists of silicon chips containing 20 microprocessors, 19 of which will behave as neurons while the remaining processor records the activities. Each chip will replicate about 20,000 human neurons, and unlike normal processors, Furber's approach will eliminate the "clocks" that typically synchronize the processes, providing a much more accurate representation of how biological neurons somehow keep on ticking (so to speak). The Spinnaker project is hoping to reach a functioning system containing 50 chips and a thousand processors within two years, and although we don't claim to be experts on the subject, it sounds like they should take a peek at how those self-healing chips are coming along, eh?[Thanks, Ian]
