The United States is back at the top of the list of the world’s 500 fastest supercomputers. The influential Top500 list, produced twice a year (and currently on its 39th edition), crowned “Sequoia,” an IBM BlueGene/Q system installed in the Department of Energy’s Livermore National Laboratory in Livermore, Calif. as the world’s fastest supercomputer. Indeed, the system can leverage its 1,572,864 processing cores to deliver 16.32 sustained petaflops (with one petaflop the equivalent of one quadrillion floating point operations per second) on the Linpack benchmark, a widely accepted performance yardstick based on the ability to solve a dense system of linear equations. And to what task does Sequoia deploy that massive amount of computing power? Researchers at the National Nuclear Security Administration (NNSA) use it to conduct virtual nuclear weapons tests, allowing them to test the country’s aging stockpile without actually setting off a nuclear bomb in violation of treaty agreements. “Computing platforms like Sequoia help the United States keep its nuclear stockpile safe, secure, and effective without the need for underground testing,” NNSA administrator Thomas D’Agostino wrote in a statement tied to the release of the list. Next on the Top500 list was Fujitsu’s “K Computer” installed at the RIKEN Advanced Institute for Computational Science (AICS) in Kobe, Japan. It delivered 10.51 petaflops, impressive but not close to Sequoia. Third on the list was the “Mira” supercomputer based at the Argonne National Laboratory in Illinois; it proved capable of 8.15 sustained petaflops on the Linpack benchmark. Like Sequoia, Mira is also an IBM BlueGene/Q system. In total, IBM was responsible for 213 systems on the list (or 42.6 percent), with Hewlett-Packard coming in second with 138 systems (27.6 percent). Cray came in third with 5.4 percent, followed by Appro with 3.6 percent, and SGI and Bull with 3.2 percent each. Fourth on the list was the “SuperMUC” supercomputer based at Leibniz Rechenzentrum in Germany. The top five was rounded out by the “Tianhe-1A” at the National Supercomputing Center in Tianjin, China. The full Top500 list is available online. It was assembled by Hans Meuer of the University of Mannheim, Germany, along with Erich Strohmaier and Horst Simon of Lawrence Berkeley National Laboratory, and Jack Dongarra of the University of Tennessee, Knoxville. In total, the systems on the list delivered roughly 123.4 petaflops’ worth of processing performance—up from November 2011, when the previous edition of the list reported a combined 74.2 petaflops. Imagine what you could do with a data center with that titanic level of processing power. Image: Jacqueline McBride/LLNL Sequoia Supercomputer Takes Crown for U.S.
The United States is back at the top of the list of the world’s 500 fastest supercomputers. The influential Top500 list, produced twice a year (and currently on its 39th edition), crowned “Sequoia,” an IBM BlueGene/Q system installed in the Department of Energy’s Livermore National Laboratory in Livermore, Calif. as the world’s fastest supercomputer. Indeed, the system can leverage its 1,572,864 processing cores to deliver 16.32 sustained petaflops (with one petaflop the equivalent of one quadrillion floating point operations per second) on the Linpack benchmark, a widely accepted performance yardstick based on the ability to solve a dense system of linear equations. And to what task does Sequoia deploy that massive amount of computing power? Researchers at the National Nuclear Security Administration (NNSA) use it to conduct virtual nuclear weapons tests, allowing them to test the country’s aging stockpile without actually setting off a nuclear bomb in violation of treaty agreements. “Computing platforms like Sequoia help the United States keep its nuclear stockpile safe, secure, and effective without the need for underground testing,” NNSA administrator Thomas D’Agostino wrote in a statement tied to the release of the list. Next on the Top500 list was Fujitsu’s “K Computer” installed at the RIKEN Advanced Institute for Computational Science (AICS) in Kobe, Japan. It delivered 10.51 petaflops, impressive but not close to Sequoia. Third on the list was the “Mira” supercomputer based at the Argonne National Laboratory in Illinois; it proved capable of 8.15 sustained petaflops on the Linpack benchmark. Like Sequoia, Mira is also an IBM BlueGene/Q system. In total, IBM was responsible for 213 systems on the list (or 42.6 percent), with Hewlett-Packard coming in second with 138 systems (27.6 percent). Cray came in third with 5.4 percent, followed by Appro with 3.6 percent, and SGI and Bull with 3.2 percent each. Fourth on the list was the “SuperMUC” supercomputer based at Leibniz Rechenzentrum in Germany. The top five was rounded out by the “Tianhe-1A” at the National Supercomputing Center in Tianjin, China. The full Top500 list is available online. It was assembled by Hans Meuer of the University of Mannheim, Germany, along with Erich Strohmaier and Horst Simon of Lawrence Berkeley National Laboratory, and Jack Dongarra of the University of Tennessee, Knoxville. In total, the systems on the list delivered roughly 123.4 petaflops’ worth of processing performance—up from November 2011, when the previous edition of the list reported a combined 74.2 petaflops. Imagine what you could do with a data center with that titanic level of processing power. Image: Jacqueline McBride/LLNL