China strengthens its position in HPC once again
As was widely expected, Tianhe-2, the supercomputer developed by China’s National University of Defence Technology, has retained its position as the world’s No. 1 system for the sixth consecutive time.
The official announcement came as part of the 46th edition of the Top500, the twice-yearly list of the world’s most powerful supercomputers, published to coincide with the US supercomputing conference SC15, held this year in Austin, Texas.
Tianhe-2, which means Milky Way-2, led the list with a performance of 33.86 Petaflops on the Linpack benchmark. This is roughly twice the performance of the second-ranking machine, Titan, a Cray XK7 system installed at the US Department of Energy’s (DOE) Oak Ridge National Laboratory. Titan, the top system in the United States and one of the most energy-efficient systems on the list, achieved 17.59 Petaflops on the Linpack benchmark.
Overall, change at the top of the list is again slower than has been seen in previous years, with only two new systems in the Top 10 - Trinity at No. 6 and Hazel-Hen at No. 8. Trinity is a Cray XC system which has 301,056 cores and which achieved 8.1 petaflops on the Linpack benchmark. Trinity is managed and operated by Los Alamos National Laboratory and Sandia National Laboratories under the Alliance for Computing at Extreme Scale (ACES) partnership. Hazel-Hen is also a Cray XC system installed in Germany at the HLRS - Höchstleistungsrechenzentrum Stuttgart - and features 185,088 cores, it achieved a benchmark score of 5.6 petaflops.
Six of the Top 10 systems were installed in 2011 or 2012, Tianhe-2 in 2013 and only Trinity, Hazel-Hen, and Shaheen II in Saudi Arabia were installed in 2015. Although this represents a slowdown of new systems entering the very top of the list, it is not quite time to lament the death of HPC as there will likely be a comparatively large number of new entries in 2016, three of which will likely come from the US, DOE’s Coral programme. As Steve Conway explained in Scientific Computing World’s yearly supplement the ‘HPC Yearbook’ movement at the very top of the Top500 list is not necessarily indicative of the entire industry.
Conway, research vice president of high-performance computing at the market-research company IDC, said: ‘In the years 2010 – 2012 particularly, there was unusually large spending on big supercomputers. The next cycle will be coming in 2016/7 and many of those procurements have already been made. The market in any one year can be thrown one way or the other by just a few, very large purchases at the top.’
If the whole list is taken into account there are some significant trends in the development of HPC systems, both in terms of geography and the technology that is being used in HPC today.
China nearly tripled the number of systems on the latest list, a huge jump from 37 systems on the last list to 109 this time around. China is now at the No. 2 for total number of systems as well as retaining its No.2 position in terms of total performance in the Top500. China is also carving out a bigger share as a manufacturer of high performance computers as grass roots efforts behind technology development begin to come to fruition in China.
The development of supercomputing in China has been the subject of much attention this year with a number of key developments including a home grown accelerator, aptly named the ‘China Accelerator’ and the development of two 100+ petaflop systems, one being the Tianhe-2.
More information on China’s developmental efforts in HPC can be found in several pieces from Scientific Computing World including; ‘China: not one but two 100 Petaflop machines within a year?’, ‘An open invitation to work on the world’s fastest computer’ and ‘Chinese Government kicks commercial companies overseas’.
Conversely, the US lost out as China moved up the list: the number of top supercomputers in the United States fell to the lowest point since the TOP500 list was created in 1993.
In a move that should not surprise anyone familiar with the development of HPC over the last few years, the total number of systems using accelerator technology rose again this year. A total of 104 systems on the list are using accelerator or co-processor technology, up from 90 on the June 2015 version of the TOP500. 66 of these use Nvidia chips, three use ATI Radeon, and there are now 27 systems with Intel Xeon Phi technology. Four systems use a combination of Nvidia and Intel Xeon Phi accelerators.