June 18, 2008
ARMONK, N.Y., June 18 -- IBM's history-making hybrid supercomputer, built for the National Nuclear Security Administration's (NNSA) Los Alamos National Lab, burned its way into the TOP500 Supercomputer record book today as the most powerful system in the world -- by a wide margin. Its sustained performance of 1.02 petaflops (1.02 quadrillion calculations per second) puts the system in a class of its own -- more than three times faster than the nearest non-IBM system.
The official results were reported today during the International Supercomputing Conference in Dresden, Germany, where the bi-annual listing of the World's TOP500 Supercomputer Sites was released.
Built by IBM for the NNSA and housed at its Los Alamos National Laboratory, the petaflop-smashing system gets its world-leading power from 12,240 IBM PowerXCell 8i Cell Broadband Engine processors -- derived from chips that power today's most popular videogame consoles. 6,562 AMD Opteron Dual-Core processors perform basic compute functions, freeing the IBM PowerXCell 8i chips for the math-intensive calculations that are their specialty.
This "hybrid" architecture, which optimizes the strength of multiple types of processors, is an IBM hallmark. The design is analogous to that of a hybrid car with similar benefits. For example, if the NNSA supercomputer were built with standard x86 chips alone, the system would have been significantly larger and would have required much more power.
While the NNSA supercomputer will be used for ensuring the reliability and safety of the nation's nuclear weapons stockpile, it also sets the pace for future research in a variety of scientific and commercial fields including biotech, alternative energy, climate change and physics. IBM expects its hybrid design to lead the way to a commercial supercomputer platform that will support new scientific research and engineering workloads unthinkable just a decade ago.
IBM Sets the Pace for TOP500
IBM continued its pace-setting leadership of the TOP500 with a trifecta showing in the top three spots and a total of 210 systems on the list -- the most of any supercomputer vendor. IBM also had the most aggregate performance on the list with 5.6 petaflops (48% of total); and the most systems in the top 10, top 50, and top 100.
The No.2 fastest computer in the world is an IBM Blue Gene/L system at NNSA's Lawrence Livermore National Lab in California, which clocked in at 478 teraflops (478 trillion calculations per second). Team Blue Gene also held the No.3 spot with a 450 teraflop performance from the Blue Gene/P system housed at the Department of Energy's Argonne National Lab in Chicago.
IBM also had the most power efficient systems: IBM QS22 PowerXCell 8i processor-based supercomputers at IBM Germany and Fraunhofer; and the NNSA system; and the fastest machine in Europe -- the Blue Gene/P at Juelich Research Centre in Germany.
The "TOP500 Supercomputer Sites" is compiled and published by supercomputing experts Jack Dongarra from the University of Tennessee; Erich Strohmaier and Horst Simon of the Department of Energy's NERSC/Lawrence Berkeley National Laboratory; and Hans Meuer of the University of Mannheim (Germany). The entire list can be viewed at http://www.top500.org/.
For more information about IBM supercomputing, visit http://www-03.ibm.com/servers/deepcomputing/.
Watch a video about Roadrunner on YouTube: http://www.youtube.com/watch?v=bpA129SHSuI.
More news about Roadrunner: http://www-03.ibm.com/press/us/en/pressrelease/24405.wss.
10/30/2013 | Cray, DDN, Mellanox, NetApp, ScaleMP, Supermicro, Xyratex | Creating data is easy… the challenge is getting it to the right place to make use of it. This paper discusses fresh solutions that can directly increase I/O efficiency, and the applications of these solutions to current, and new technology infrastructures.
10/01/2013 | IBM | A new trend is developing in the HPC space that is also affecting enterprise computing productivity with the arrival of “ultra-dense” hyper-scale servers.
Ken Claffey, SVP and General Manager at Xyratex, presents ClusterStor at the Vendor Showdown at ISC13 in Leipzig, Germany.
Join HPCwire Editor Nicole Hemsoth and Dr. David Bader from Georgia Tech as they take center stage on opening night at Atlanta's first Big Data Kick Off Week, filmed in front of a live audience. Nicole and David look at the evolution of HPC, today's big data challenges, discuss real world solutions, and reveal their predictions. Exactly what does the future holds for HPC?