Highlights - June 2013

Tianhe-2 (Milky Way-2), a system developed by China’s National University of Defense Technology (NUDT) and to be deployed at the National Supercomputer Center in Guangzho, China is the new No. 1 system with 33.86 Petaflop/s on the Linpack benchmark. The system currently has 16,000 nodes each with two Intel Xeon IvyBridge processors and three Xeon Phi processors for a combined total of 3,120,000 computing cores. It features number of Chinese-developed components, including the TH Express-2 interconnect network, front-end processors, operating system, and software tools.

The Tianhe-2 uses the Kylin Linux operating system. Officially approved for use in 2006, Kylin was developed by the National University for Defense Technology, is compatible with other mainstream operating systems, and supports multiple microprocessors and computers of different architectures. In addition, NUDT developed OpenMC, a directive-based intra-node programming model, similar to Open-MP and either CUDA, OpenACC, or OpenCL.

The Tianhe-2 has a front-end system composed of 4,096 Galaxy FT-1500 CPUs, designed and developed at NUDT. The FT-1500 is 16 cores and based on SparcV9. Its performance is 144 Gflop/s and each chip runs at 65 watts. By comparison the Intel Ivy Bridge has 12 cores with a peak performance of 211 Gflop/s. The power consumption of Tianhe-2 while running Linpack was 17.8 MW.

Other highlights from the Top 10:

  • Titan, a Cray XK7 system installed at the Department of Energy’s (DOE) Oak Ridge National Laboratory and was previously the No. 1 system is now the No. 2 system. It achieved 17.59 petaflop/s on the Linpack benchmark using 261,632 of its NVIDIA K20x accelerator cores. Titan is one of the most energy efficient systems on the list consuming a total of 8.21 MW and delivering 2,143 MFlops/W.
  • Sequoia, an IBM BlueGene/Q system installed at DOE’s Lawrence Livermore National Laboratory, is now the No. 3 system. It was first delivered in 2011 and has now achieved 17.17 petaflop/s on the Linpack benchmark using 1,572,864 cores. Sequoia is also one of the most energy efficient systems on the list consuming a total of 7.84 MW and delivering 2,176.6 MFlops/W.
  • Fujitsu’s K computer installed at the RIKEN Advanced Institute for Computational Science (AICS) in Kobe, Japan, is now the No. 4 system with10.51 Pflop/s on the Linpack benchmark using 705,024 SPARC64 processing cores.
  • A second BlueGene/Q system (Mira) installed at DOE’s Argonne National Laboratory is No. 5 with now 8.59 Petaflop/s on the Linpack benchmark using 786,432 cores.
  • At No. 6 is now the upgraded Stampede, a Dell PowerEdge C8220 system installed at the Texas Advanced Computing Center of the University of Texas, Austin. It also uses Intel Xeon Phi processors (previously known as MIC) to achieve its 5.17 PFlop/s.
  • The most powerful system in Europe dropped to No. 7. It is also a BlueGene/Q system called JUQEEN installed at the Forschungszentrum Juelich in Germany and is listed with 5.01 PFlop/s.
  • No. 8 is taken by another IBM BlueGene/Q system at Lawrence Livermore National Laboratory. It was temporarily combined with the No. 3 system but is now operated independently. It achieved 4.29 PFlop/s.
  • At No. 9 is the second system in Europe, the SuperMUC, an IBM iDataplex system with Intel Sandybridge processors installed at Leibniz Rechenzentrum in Germany with 2.9 PFlop/s.
  • The Chinese Tianhe-1A system, the No. 1 on the TOP500 in November 2010, is now No. 10 with 2.57 Pflop/s Linpack performance. 

Highlights from the Overall List

  • There are 26 systems with performance greater than a petaflop/s on the list up, from 23 six months ago.
  • The new No. 1 system, Tianhe-2, and the No. 6 system, Stampede, are using Intel Xeon Phi processors to speed up their computational rate. The No. 2 system Titan and the No. 10 system Tianhe-1A are using NVIDIA GPUs to accelerate computation.
  • A total of 54 systems on the list are using Accelerator/Co-Processor technology, down from 62 in November 2012. Thirty-nine of these use NVIDIA chips, three use ATI Radeon, and there are now 11 systems with Intel MIC technology (Xeon Phi).
  • Intel continues to provide the processors for the largest share (80.6 percent) of TOP500 systems.
  • Eighty-eight percent of the systems use processors with six or more cores and 67 percent with eight or more cores.
  • IBM’s BlueGene/Q is still the most popular system in the TOP10 with 4 entries including the No. 3, 5, 7, and No. 8.
  • The number of systems installed in China has now stabilized at 66 with 72 and 68 on the last two lists.  China has occupied the No. 2 position as a user of HPC, ahead of Japan, UK, France, and Germany. Due to Tianhe-2 China now also took the No. 2 position in the performance share ahead of Japan.

General highlights from the TOP500 since the November 2012 edition:

  • The entry level to the list moved up to the 96.6 Tflop/s mark on the Linpack benchmark, compared to 76.5 Tflop/s six months ago.
  • The last system on the newest list was listed at position 322 in the previous TOP500.
  • Total combined performance of all 500 systems has grown to 223 Pflop/s, compared to 162 Pflop/s six months ago and 123 Pflop/s one year ago.
  • The entry point for the TOP100 increased in six months from 244 Tflop/s to 290 Tflop/s.
  • The average concurrency level in the TOP500 is 38,700 cores per system, up from 29,796 six months ago and 26,327 one year ago. 

Vendor Trends

  • A total of 403 systems (80 percent) are now using Intel processors, slightly up from 76 percent six months ago.
  • Intel is followed by the AMD Opteron family with 50 systems (10 percent), slightly down from 12 percent on the previous list.
  • The share of IBM Power processors is at 42 systems (8.4 percent).
  • InfiniBand technology is now found on 203 systems, down from 223 systems, and is the most-used internal system interconnect technology. Gigabit Ethernet climbed back up to 216 systems from 189 systems, in large part thanks to 75 systems now using 10G interfaces.
  • IBM and Hewlett-Packard continue to sell the bulk of the systems at all performance levels of the TOP500.
  • HP won the lead in systems and now has 188 systems (38 percent) compared to IBM with 160 systems (32 percent). HP is up from 146 systems (29.2 percent) six months ago, compared to IBM with 193 systems (38.6 percent) six months agoIn the system category, Cray is third with 10 percent. 

Performance Trends

  • IBM remains the clear leader in the TOP500 list in performance and has a considerable lead with a 33 percent share of installed total performance (down from 41 percent).
  • Thanks to Tianhe-2 and Tianhe-1A, NUDT contributes 16.8% performance to the list.
  • Cray’s share in performance is now at 15.3 percent down from 17.4 percent.
  • HP is now fourth even so it increased its share to 14.1 percent up from 11.4 percent. 

Geographical observations

  • The U.S. is clearly the leading consumer of HPC systems with 252 of the 500 systems (250 last time). The European share (112 systems compared to 105 last time) is still lower than the Asian share (119 systems, down from 123 last time).
  • Dominant countries in Asia are China with 66 systems (down from 72) and Japan with 30 systems (down from 32).
  • In Europe, UK, France, and Germany, are almost equal with 29, 23, and 19 respectively.