Supercomputer technology has been a major concern for different technology developers since the 1960s, from early room sized computers through to advanced multi core processors and specialist equipment used for high level tasks and their server capacity. Today’s supercomputers are regularly ranked by the online Top500 List, with the most significant competition tending to come between Japan, China and the United States. Most operate according to their petraflop speeds, use Linux as an operating system, and are able to be used for high level processing solutions for astrophysics and engineering. The current top 6 supercomputers in the world include:
6 – TSUBAME 2.0 – HP ProLiant SL390s G7 Xeon 6C X5670, Nvidia GPU, Linux/Windows
Tokyo based, the HP ProLiant has been a collaboration between the GSIC Centre and NEC. The computer has a performance rate of 2,288 Tflops, and also uses NVIDIA Tesla processors. The computer builds on HP ProLiant’s work with Compaq to speed up performance, and to optimize multiple processors and a super advanced cache and memory.
5 – Nebulae – Dawning TC3600 Blade, Intel X5650, NVidia Tesla C2050 GPU
This Chinese computer is the first supercomputer to use blade technology, and can achieve 13 petraflops of speed, while also being theoretically possible of going much higher. The result of research into Dawning Cluster blade technology, the computer was launched in 2010, and features Nvidia Tesla and Intel X5650 processors. Set to be the one of the important next generation computers for the Chinese, the use of blade servers is notable for its high power efficiency, and maximisation of space through multiple components combining together different blade servers.
4 – Jaguar – Cray XT5-HE Opteron 6-core 2.6 GHz
The Cray XT5-HE, or Jaguar, represents an American addition to the list, and has been ind development at Cray Inc. in Tennessee since 2009. The computer can achieve 1,759 teraflops, and uses Opteron processor cores. It also uses a special version of Linux, which is optimised for Cray’s hardware. The Jaguar has previously been ranked as the fastest computer in the world, and is also defined by its Lustre file systems, and by its use in astrophysics and chemistry calculations.
3 – Tianhe-1A – NUDT YH MPP, Xeon X5670 6C 2.93 GHZ, VIDIA 2050
The Tianhe 1a supercomputer is one of China’s key contributions to supercomputer technology, and is primarily used as a research and development device. Containing 229376 GB of memory, the NUDT runs on Linux, and is based in Tianjin. Also features Xeon technology.
2 – K Computer, SPARC64 VIIIfx 2.0 GHz, Tofu interconnect
Until 2011, the K computer was the worlds fastest supercomputer at 10.51 petaflops. This supercomputer is a Japanese project that has been developed by the RIKEN Advanced Science Institute and Fujitsu, and can achieve 8 petraflops. Setting LINPACK records, the K includes 8 core SPARC64 VIIIfx processors, and includes 864 cabinets. The power needed to run the computer is equivalent to 10,000 suburban homes, or 1 million linked desktop computers. The iconic status of the K in Japan has even led a local rapid transit system to name a line after its achievements.
1 – IBM Sequoia Blue Gene/Q system
IBM Sequoia is a petascale Blue Gene/Q supercomputer developed by IBM. Sequoia replaced the K computer as the world’s fastest supercomputer, with a LINPACK performance of 16.32 petaflops, 55% faster than the K computer’s 10.51 petaflops, using 123% more sockets than the K computer’s 705,024 sockets. Sequoia is also more energy efficient, as it consumes 7.9 MW, 37% less than the K computer’s 12.6 MW. The entire supercomputer runs on Linux, with Compute Node Linux running on over 98,000 nodes, and Red Hat Enterprise Linux running on 768 I/O nodes that are connected to the file system.