China’s National University of Defense Technology set a new performance record for these super machines. Built by the National University of Defense Technology, located at the National Supercomputing Center in Tianjin, a supercomputer was revealed on Thursday at the annual meeting of National High Performance Computing (HPC China 2010) in Beijing.
The machine, called Tianhe-1A (Tianhe means “Milky Way”), has more than 2.5 petaflops per second of peak processing power, chipmaker Nvidia says. Nvidia supplied graphics processors to Tianhe-1A. That means the Chinese computer is far more powerful than the previous record holder, the Cray XT5 Jaguar. The Cray, which is housed at the Oak Ridge National Laboratory in Tennessee, has a peak output of 1.75 petaflops per second.
Tianhe-1A epitomizes modern heterogeneous computing by coupling massively parallel GPUs with multi-core CPUs, enabling significant achievements in performance, size and power, said Nvidia. Furthermore, the system uses 7,168 Nvidia Tesla M2050 GPUs and 14,336 CPUs as opposed to requiring more than 50,000 CPUs and twice as much floor space to deliver the same performance using CPUs alone, argued Nvidia.
US Scientists have confirmed that the Tianhe supercomputer boasts a computing speed which is 1.4% faster than Cray XT5 Jaguar, which was the former leading supercomputer in Oak Ridge, Tennessee. The former top supercomputer from US had scooped the leading spot in June this year with a computational speed of 1.75 petaflops, which is 1750 trillion calculations per second.
The Tianhe-1A is currently held in Tianjin, a city at the north port next to Beijing, and is a creation of the National University of Defence Technology. Although a making of Chinese intellectuals, integrated into the framework of the Tianhe is a myriad of numerous US-made processing chips such as Nvidia and Intel. The processors are housed in over 100 cabinets, the size of a refrigerator, weighing over 155 tons.
“GPUs are redefining high performance computing,” said Jen-Hsun Huang, president and CEO of Nvidia. “With the Tianhe-1A, GPUs now power two of the top three fastest computers in the world today. These GPU supercomputers are essential tools for scientists looking to turbocharge their rate of discovery.” Fully operational, the Tianhe-1A supercomputer will be operated as an open access system for large-scale scientific computations.