Cray blows by IBM to regain supercomputing crown
A Cray supercomputer at Oak Ridge National Laboratory has regained the title of the world's most powerful supercomputer, overtaking the installation that was ranked at the top in June, while China entered the Top 10 with a hybrid Intel-AMD system.
The upgraded Jaguar supercomputer at Oak Ridge, in Tennessee, now boasts a speed of 1.759 petaflops from its 224,162 cores, while the IBM Roadrunner system at the U.S. Department of Energy's Los Alamos National Laboratory in New Mexico slowed slightly to 1.042 petaflops after it was repartitioned. A petaflop is one thousand trillion calculations per second.
The list of the Top 500 supercomputers, set to be released on Monday during the SC09 supercomputing conference in Portland, Oregon, is compiled twice a year and is now in its 34th installment. The total capacity of the systems on the new list is 27.6 petaflops, up from 22.6 petaflops on the previous list in June.
Roadrunner debuted in June 2008 as the first computer to surpass 1 petaflop on the Linpack benchmark test used to rank systems in the Top 500. It held the top spot in June 2009 with 1.105 petaflops, but lost its place after being repartitioned. Jaguar, which was in second place in June with 1.059 petaflops, was upgraded with new processors and surged ahead to take the lead. It is based on the Cray XT5 Linux supercomputer platform, which uses Advanced Micro Devices Opteron (AMD) processors. Its total peak capability is 2.3 petaflops.
The No. 3 system is Kraken, at the National Institute for Computational Sciences at the University of Tennessee, which performs at 832 teraflops. This Cray XT5 supercomputer was ranked No. 6 in June, when it was rated at just 463 teraflops.
China's fastest supercomputer ever, the Tianhe-1 in the city of Tianjin, achieved 563 teraflops for the No. 5 ranking. It uses Intel Xeon processors with Advanced Micro Devices GPUs (graphics processing units) as accelerators. Each node of the 71,680-core system has two Xeons attached to two AMD GPUs, according to the compilers of the Top 500 list. Tianhe-1 was built by the National University of Defense Technology for the National SuperComputer Center and is intended to provide high-performance computing services in northeastern China. Applications will include petroleum exploration and aircraft design.
The only other Top 10 system outside the U.S. was Jugene, built by IBM at the Forschungszentrum Juelich in Germany, which was ranked No. 4. U.S. computers dominated the Top 500 overall, making up 277 of the systems, with Europe accounting for 153 and Asia for 50. Just to make it onto the new Top 500 list, a computer needed to achieve at least 20 teraflops, up from 17.1 teraflops earlier this year.
Intel processors power 402 of the systems on the list, or 80.4 percent, up slightly from 399 in June. The IBM Power architecture is the second most commonly used, with 52 systems, down from 55. AMD's Opteron family appears in 42 of the systems.
Most of the Top 500 supercomputers -- 426 systems -- now use quad-core processors. Only 59 use dual-core chips, and just four systems are based on single-core architectures. There were six systems on the latest list using IBM's nine-core Cell Broadband Engine processor, also used in the PlayStation 3. Gigabit Ethernet is the internal interconnect technology in 259 installations, compared with 181 using InfiniBand.
Hewlett-Packard led in the number of systems on the list, with 210 supercomputers or 42 percent, compared with 185 for IBM. However, the IBM systems accounted for the most computing power, with 34.8 percent of total performance, down from 39.8 percent. HP held 22.8 percent.
The Top 500 list is compiled by Hans Meuer of the University of Mannheim in Germany, Jack Dongarra of the University of Tennessee at Knoxville, and Erich Strohmaier and Horst Simon of the National Energy Research Scientific Computing Center at Lawrence Berkeley National Laboratory in California.