Chinese computing power has leapfrogged Europe and Japan, with a new supercomputer ranked as the world’s second fastest.
The Tianhe-1 – the name means Dawning Nebulae – is sited at the National Supercomputing Center in Shenzhen. Based on chips from Intel and Nvidia, it’s achieved a sustained computing speed of 1.27 petaflops in the Top 500 supercomputer ranking.
The top spot is still held by the Cray Jaguar supercomputer, which is based at the Oak Ridge National Laboratory in Tennessee, which has a speed of 1.75 petaflops. It’s used mainly for nuclear weapons simulations, as well as physics and climate research.
But this may be the last year in which the Jaguar keeps that honor. In deed, it could be argued that it’s lost it already, as the Chinese machine has a higher theoretical speed limit.
And the Chinese are also working on a new supercomputer, based on a domestic microsprocessor, which is expected later this year. It’s entirely likely thast this will break the speed record.
Nevertheless, the US still dominates the list with more than half the top 500 machines. And it, too, is working on new supercomputing technology, aiming to outdo today’s performance by a thousand times within the next ten years.
AMD reckons it’s the microprocessor winner – the Jaguar is based on the six-core AMD Opteron chip and features almost a quarter of a million cores. Another 50-odd of the top 500 are also based on its processors.
“Our customers are selecting AMD platforms for supercomputing because they provide the cores, the memory, the power savings and clearly the performance that the world’s leading research institutions require for their ground-breaking work,” says John Fruehe, director of server and embedded product marketing.
The full list is available here.