China’s Tianhe-2 Remains Atop Fastest Supercomputer List

There was little change on the top end of the twice yearly Top500 list of the world’s fastest supercomputers. Only two new systems made it into the top 10.

Tianhe-2, housed at China’s National University of Defense Technology, has topped the list since 2013. It’s powered by Intel’s 12-core Xeon E5-2692 processors and Xeon Phi coprocessors, and offers performance of 33.86 petaflops (quadrillions of calculations per second).

Titan, housed at the Oak Ridge National Laboratories in the United States, remained in second place with a performance of almost 17.6 petaflops. The system is based on Cray’s XK7 servers, runs on 16-core Opteron 6274 processors from Advanced Micro Devices and uses Nvidia’s Tesla K20x GPU accelerators.

Sequoia is one of two IBM systems at the top of the list. It’s a Blue Gene/Q supercomputer installed at the Lawrence Livermore Labs and provides almost 17.2 petaflops of performance.

The K computer at Japan’s Riken Advanced Institute for Computational Science was ranked No. 1 in 2011, and became the first system to break the 10-petaflop barrier. It runs on Fujitsu’s SPARC64 VIIIfx chips and comes in with 10.5-petaflop performance.

Mira, installed at Argonne National Lab in the United States, is the second IBM system in the top 10. The supercomputer provides a performance of more than 8.5 petaflops.

Trinity, a Cray XC40 system, is operated by Los Alamos National Labs and Sandia National Labs as part of the Alliance for Computing at Extreme Scale (ACES) partnership. It is powered by Intel’s 16-core Xeon E5-2698v3 chips, has 301,056 cores and hit 8.1 petaflops of performance.

Piz Daint, installed at the Swiss National Computing Center, is one of five Cray supercomputers in the top 10. It’s based on Cray’s XC30 system and is powered by eight-core Xeon E5-2670 processors and uses Nvidia’s Tesla K20x GPU accelerators.

The system is based on Cray’s XC40 systems that run 12-core Xeon E5-2680v3 processors. It has 185,088 cores and a performance of 5.6 petaflops.

Shaheen II, a Cray XC40 system powered by 16-core Xeon E5-2698v3 processors and with 5.5-petaflop performance, is installed at the King Abdullah University of Science and Technology in Saudi Arabia.

Stampede is a supercomputer installed at the Texas Advanced Computing Center at the University of Texas. The system is based on Dell’s PowerEdge C8220 servers and uses Intel’s Xeon E5-2680 chips and Xeon Phi SE10P coprocessors. It offers 5.1-petaflop performance.