For the third time running, Tianhe-2, a supercomputer developed by China's National University of Defence Technology, has held onto the top spot as the world's most powerful supercomputer.
Tianhe-2 reports performance of 33.86 petaflop/s (quadrillions of calculations per second) on the Linpack benchmark in the latest edition of the Top500 list of the world's most powerful supercomputers.
Supercomputers tend to be used for projects requiring huge amounts of parallel processing, such as genome modelling, weather forecasting or oil and gas exploration. There's also increasing interest in supercomputing technology from enterprises as big data analytics becomes more prevalent.
There was little change in the ranking of the world's top 10 supercomputers in the latest edition of the Top500 list, with the only new entry at number 10 — a 3.14 petaflop/s Cray XC30 installed at an 'undisclosed' US government site.
And it's not just at the top of the list the China's supercomputing power is shown. Although the US remains the top country by supercomputers with 233, this is down from 265 on the November 2013 list. In contrast, the number of Chinese systems on the list rose from 63 to 76, giving China nearly as many supercomputers as the UK (30), France (27) and Germany (23) combined. Japan also increased its showing, up to 30 from 28 on the previous list.
The Top500 list ranks systems by performance running the same Linpack benchmark application, but for the second consecutive list, the overall growth rate of all the systems is at an "historical low" — largely because of a lack of new, very large systems that might appear near the top of the list.
Total combined performance of all 500 systems has grown to 274 Pflop/s, compared to 250 Pflop/s six months ago and 223 Pflop/s one year ago which was described as a "noticeable slowdown in growth compared to the previous long-term trend".
The list also notes that the top-ranked system, Tianhe-2, and the seventh-placed system, Stampede, use Intel Xeon Phi processors to speed up their computational rate. The runner-up, Titan, and the number six system, Piz Daint, use NVIDIA GPUs to accelerate computation. A total of 62 systems on the list are using accelerator/coprocessor technology, up from 53 from November 2013. Forty-four of these use NVIDIA chips, two use ATI Radeon and there are now 17 systems with Intel MIC technology — Xeon Phi.
The average number of accelerator cores for these 62 systems is 78,127 cores/system. Intel continues to provide the processors for the largest share (85.4 percent) of Top500 systems. The share of IBM Power processors remains at eight percent, while the AMD Opteron family is used in six percent of the systems — down from nine percent in the previous list.
HP accounts for 182 systems (36 percent) compared to IBM with 176 systems (35 percent). HP had 196 systems (39 percent) six months ago, while IBM had 164 systems (33 percent) six months ago. In the system category, Cray remains third with 10 percent (51 systems).