IBM's Roadrunner is five times faster than the Chinese computer, but cost more than three times as much



China has built the world's seventh-fastest super computer, in another sign of its ambition to be a leading force in high-technology.

The Communist Party believes super computers are essential to China's efforts in science and advanced design.

The Dawning was finished too late to be included in the official rankings of the world's most powerful machines, which was released last week. However, the Dawning is thought to be only slightly slower than the sixth-ranked IBM Blue Gene/P computer at the Juelich Research Centre in Germany.


More . . .

China's first super computer off the assembly line in Tianjin

On the morning of September 16, China's first high performance supercomputer Shuguang 5000A came off the assembly line in Tianjin.

"Shuguang 5000A not only makes China the second country after the United States that has the ability to research, develop, produce and use supercomputer with computing speed over 100 trillion times per second, but also greatly enhance China's technological competitiveness and overall national strength.

Shuguang 5000A can reach peak computing speed of 230 trillion times per second.


More . . .

A Chinese supercomputer has ranked as the world’s second fastest machine, surpassing European and Japanese systems and underscoring China’s aggressive commitment to science and technology.

The Dawning Nebulae, based at the National Supercomputing Center in Shenzhen, China, has achieved a sustained computing speed of 1.27 petaflops — the equivalent of one thousand trillion mathematical operations per second — in the twice annual ranking of the world’s fastest 500 hundred computer.

There had been some expectation that they would make an effort to complete a system based on Chinese designed components in time for the June ranking. The Nebulae is based on chips from Intel and Nvidia.

The new system, which is based on a microprocessor that has been designed and manufactured in China, is now expected later this year. A number of supercomputing industry scientists and engineers said that it is possible that the new machine will claim the title of world’s fastest.



More . . .

I would LOVE to boinc with that computer!

Thanks for all the interesting news you put up on the forums, Sid2. I really enjoy them.
Rod

China's NSCS [National Supercomputing Center in Shenzhen] is enthusiastically adopting Dawning's TC3600 blade servers, equipped with Intel's six-core X5650 processors and Nvidia's C2050 GPUs. The exact configuration of the Nebulae machine at NSCS was not available at press time, but the TC3600 blade server is a 10U chassis that holds ten two-socket blades. The C2050s are PCI-Express GPU co-processors with 448 cores and 3 GB of their own GDDR5 memory, rated at 515 gigaflops doing double-precision floating point math and 1.03 teraflops doing single-precision.

The Top 500 ranking for Nebulae does not provide blade or GPU count, but the word on the street is that it has 4,700 nodes. What the Top 500 does say the machine has 120,640 cores in total for a peak theoretical performance of 2.98 petaflops and 1.27 petaflops sustained running the Linpack test. All of the nodes in the Dawning blade cluster are linked by quad data rate (40 Gb/sec) InfiniBand switches.

The first thing to notice about the Jaguar and Nebulae supers is the difference between peak and sustained performance. For the Cray Jaguar Opteron cluster, 75.5 per cent of the flops contained in the box end up doing real Linpack work, while on the Dawning Xeon-Tesla hybrid, only 42.6 per cent of the peak performance embodied in the CPUs and GPUs actually push Linpack math. So it would seem that the all-X64 machine has the edge, right? Wrong. Jaguar cost around $200m to build and burns around 7 megawatts of juice, while the Nebulae machine probably costs on the order of $50m (that's an El Reg estimate) and burns only 2.55 megawatts of juice.


More . . .