China retains supercomputing crown as U.S. representation lingers near historic lows

A supercomputer developed by China’s National Defense University remains the fastest publically known computer in the world while the U.S. is close to an historic low in the latest edition of the closely followed Top 500 supercomputer ranking, which was published on Monday.

The Tianhe-2 computer, based at the National Super Computer Center in Guangzhou, has been on the top of the list for more than two years and its maximum achieved performance of 33,863 teraflops per second is almost double that of the U.S. Department of Energy’s Cray Titan supercomputer, which is at Oak Ridge National Laboratory in Tennessee.

The IBM Sequoia computer at Lawrence Livermore National Laboratory in California is the third fastest machine, and fourth on the list is the Fujitsu K computer at Japan’s Advanced Institute for Computational Science. The only new machine to enter the top 10 is the Shaheen II computer of King Abdullah University of Science and Technology in Saudi Arabia, which is ranked seventh.

The Top 500 list, published twice a year to coincide with supercomputer conferences, is closely watched as an indicator of the status of development and investment in high-performance computing around the world. It also provides insights into what technologies are popular among organizations building these machines, but participation is voluntary. It’s quite possible a number of secret supercomputers exist that are not counted in the list.

With 231 machines in the Top 500 list, the U.S. remains the top country in terms of the number of supercomputers, but that’s close to the all-time low of 226 hit in mid-2002. That was right about the time that China began appearing on the list. It rose to claim 76 machines this time last year, but the latest count has China at 37 computers.

While there are few major changes in the top positions in the ranking, the aggregate computing power of the 500 companies continues to advance, but the pace is slowing. The current list represents 361 petaflops per second of performance, up 31 percent on this time last year, but a noticeable slowdown in growth, according to the authors of the study.

The rise of the use of graphics processors, so-called GPU computing, is reflected in the top 10. Two machines used Nvidia K20x processors: the second-ranked Cray Titan and sixth-ranked Cray Piz Daint, which is installed at the Swiss National Supercomputing Centre.

But Intel’s Xeon E5 chip continues to outrank all others. Taken together, three generations of the chip (SandyBridge, IvyBridge and Haswell) are in 80 percent of systems, representing 67 percent of total performance.

The Top 500 list is compiled by supercomputing experts at the University of Mannheim, Germany; the University of Tennessee, Knoxville; and the Department of Energy’s Lawrence Berkeley National Laboratory.

via China retains supercomputing crown as U.S. representation lingers near historic lows | PCWorld.

World’s top supercomputer from ‘09 is now obsolete, will be dismantled

Five years ago, an IBM-built supercomputer designed to model the decay of the US nuclear weapons arsenal was clocked at speeds no computer in the history of Earth had ever reached. At more than one quadrillion floating point operations per second (that’s a million billion, or a “petaflop”), the aptly-named Roadrunner was so far ahead of the competition that it earned the #1 slot on the Top 500 supercomputer list in June 2008, November 2008, and one last time in June 2009.
Today, that computer has been declared obsolete and it’s being taken offline. Based at the US Department of Energy’s Los Alamos National Laboratory in New Mexico, Roadrunner will be studied for a while and then ultimately dismantled. While the computer is still one of the 22 fastest in the world, it isn’t energy-efficient enough to make the power bill worth it.
“During its five operational years, Roadrunner, part of the National Nuclear Security Administration’s Advanced Simulation and Computing (ASC) program to provide key computer simulations for the Stockpile Stewardship Program, was a workhorse system providing computing power for stewardship of the US nuclear deterrent, and in its early shakedown phase, a wide variety of unclassified science,” Los Alamos lab said in an announcement Friday.
Costing more than $120 million, Roadrunner’s 296 server racks covering 6,000 square feet were connected with InfiniBand and contained 122,400 processor cores. The hybrid architecture used IBM PowerXCell 8i CPUs (an enhanced version of the Sony PlayStation 3 processor) and AMD Opteron dual-core processors. The AMD processors handled basic tasks, with the Cell CPUs “taking on the most computationally intense parts of a calculation—thus acting as a computational accelerator,” Los Alamos wrote.

Los Alamos National Laboratory
“Although other hybrid computers existed, none were at the supercomputing scale,” Los Alamos said. “Many doubted that a hybrid supercomputer could work, so for Los Alamos and IBM, Roadrunner was a leap of faith… As part of its Stockpile Stewardship work, Roadrunner took on a difficult, long-standing gap in understanding of energy flow in a weapon and its relation to weapon yield.”
Roadrunner lost its world’s-fastest title in November 2009 to Jaguar, another Department of Energy supercomputer combining AMD Opterons with Cray processors. Jaguar hit 1.76 petaflops to take the title, and it still exists as part of an even newer cluster called Titan. Titan took the top spot in the November 2012 supercomputers list with a speed of 17.6 petaflops.
Supercomputing researchers are now looking toward exascale speeds—1,000 times faster than a petaflop—but major advances in energy efficiency and price-performance are necessary.
Petaflop machines aren’t automatically obsolete—a petaflop is still speedy enough to crack the top 25 fastest supercomputers. Roadrunner is thus still capable of performing scientific work at mind-boggling speeds, but has been surpassed by competitors in terms of energy efficiency. For example, in the November 2012 ratings Roadrunner required 2,345 kilowatts to hit 1.042 petaflops and a world ranking of #22. The supercomputer at #21 required only 1,177 kilowatts, and #23 (clocked at 1.035 petaflops) required just 493 kilowatts.
“Future supercomputers will need to improve on Roadrunner’s energy efficiency to make the power bill affordable,” Los Alamos wrote. “Future supercomputers will also need new solutions for handling and storing the vast amounts of data involved in such massive calculations.”
After Roadrunner is shut off today, researchers will spend a month doing experiments on “operating system memory compression techniques for an ASC relevant application, and optimized data routing to help guide the design of future capacity cluster computers.” After that, the cluster will finally be dismantled, the end of the world’s first petaflop supercomputer.
via World’s top supercomputer from ‘09 is now obsolete, will be dismantled | Ars Technica.