INDUSTRY

By ELLEN SIMON, The Associated Press — IBM Corp. claimed unofficial bragging rights Tuesday as owner of the world's fastest supercomputer. For three years running, the fastest supercomputer has been NEC's Earth Simulator in Japan. "The fact that non-U.S. vendor like NEC had the fastest computer was seen as a big challenge for U.S. computer industry," said Horst Simon, director of the supercomputing center at Lawrence Berkeley National Lab in California. "That an American vendor and an American application has won back the No. 1 spot — that's the main significance of this." Earth Simulator can sustain speeds of 35.86 teraflops. IBM said its still-unfinished BlueGene/L System, named for its ability to model the folding of human proteins, can sustain speeds of 36 teraflops. A teraflop is 1 trillion calculations per second. Lawrence Livermore National Laboratory plans to install the Blue Gene/L system next year with 130,000 processors and 64 racks, half a tennis court in size. The labs will use it for modeling the behavior and aging of high explosives, astrophysics, cosmology and basic science, lab spokesman Bob Hirschfeld said. The prototype for which IBM claimed the speed record is located in Rochester, Minn., has 16,250 processors and takes up eight racks of space. While IBM's speed sets a new benchmark, the official list of the world's fastest supercomputers will not be released until November. A handful of scientists who audit the computers' reported speeds publish them on Top500.org. Supercomputing is significant because of its implications for national security as well as such fields as global climate modeling, astrophysics and genetic research. Supercomputing technology IBM introduced a decade ago has evolved into a $3 billion to $4 billion business for the company, said Simon. Unlike the more specialized architecture of the Japanese supercomputer, IBM's BlueGene/L uses a derivative of commercially available off-the-shelf processors. It also uses an unusually large number of them. The resulting computer is smaller and cooler than other supercomputers, reducing its running costs, said Hirschfeld. He did not have a dollar figure for how much lower Blue Gene's costs will be than other supercomputers. However, other supercomputers can do things Blue Gene cannot, such as produce 3-D simulations of nuclear explosions, Hirschfeld said. ==================================================== Here are further details about IBM's Blue Gene Supercomputer project: An IBM BlueGene/L supercomputer has become the world's most powerful supercomputer, surpassing NEC's Earth Simulator in internal testing using the industry-standard LINPACK benchmark. The system was assembled by IBM engineers and researchers in Rochester, Minnesota. Designed as a new approach to cost effective, ultra-powerful supercomputing, the BlueGene/L system is also a fraction of the size and power consumption of Earth Simulator. Yet, this is only a glimpse of Blue Gene/L's full potential. The largest planned Blue Gene/L machine, which is being developed in collaboration with the Lawrence Livermore National Laboratory in California, will occupying 64 full racks, with a peak performance of 360 teraflops. When completed in 2005, IBM expects this system to lead the Top500 supercomputer list. Compared to today's fastest supercomputers, it will be more than eight times faster, consume 1/15th the power per computation and be 10 times more compact. Blue Gene/L is designed as a multipurpose system, drawing widespread interest. For instance, IBM is also working with ASTRON, a leading astronomical organization in the Netherlands, to build a 34 teraflop Blue Gene/L supercomputer. The Argonne National Laboratory in Illinois has also announced that it has committed funding to acquire a Blue Gene/L system of its own. And university collaborators have flocked to Blue Gene/L 'user group' meetings at national lab sites, including a recent meeting at Argonne, in order to learn how to use the system in their research. The Blue Gene/L prototype machines are roughly 1/20th the physical size of machines of comparable compute power that exist today -- such as Linux clusters. By making dramatic reductions in power consumption, cost and space requirements, IBM researchers are helping to turn massively parallel computing into an affordable, practical and accessible tool for science and industry. The Blue Gene Family: The Future of Supercomputing Blue Gene is an IBM supercomputing project with two overall goals: The first is to build a new family of supercomputers optimized for bandwidth, scalability and the ability to handle large amounts of data while consuming a fraction of the power and floor space required by today's fastest systems. The second is to use this computing platform to attack a broad range of challenging scientific and data analysis problems. Among the first applications IBM is exploring to harness Blue Gene's massive computing power is to model the folding of human proteins - a technique expected to give medical researchers better understanding of diseases, as well as potential cures. The first machine in the family, Blue Gene/L, is expected to operate at a peak performance of about 360 teraflops (360 trillion operations per second), and occupy 64 racks -- taking up only about the same space as half of a tennis court. Researchers at the Lawrence Livermore National Laboratory (LLNL) plan to use Blue Gene/L to simulate physical phenomena that require computational capability much greater than presently available, such as cosmology and the behavior of stellar binary pairs, laser-plasma interactions, and the behavior and aging of high explosives. ASTRON and IBM scientists plan to use IBM's Blue Gene/L supercomputer technology as the basis for a new type of radio telescope capable of looking back billions of years in time. This joint research project in high data volume supercomputing will help astronomers examine the beginnings of the earliest stars and galaxies after the formation of the universe, known as the Big Bang. The architecture is proving to be readily adaptable to a range of applications, and will be more affordable than current supercomputing resources due to its smaller physical size and power efficiency. IBM and its partners are currently exploring a growing list of applications including hydrodynamics, quantum chemistry, molecular dynamics, climate modeling and financial modeling.