14 DAY TRIAL // Given the fact that the SC'10 supercomputing conference is taking place in New Orleans as we speak, we've come across plenty of interesting news regarding this highly-specialized field, the latest pertaining to NVIDIA's, who's just revealed the fact that its Tesla GPUs power three of the world's fastest 5 supercomputers.
According to NVIDIA, Tesla GPUs were featured in the number one, three and four slots of the November 2010 list of the "Top500" fastest supercomputers in the world, with the recently announced Tianhe-1A system taking the top spot with a performance record of 2.507 petaflops.
Next in line comes the 1.75-petaflop Jaguar supercomputer (USA), the 1.27 PF Nebulae from China, the Tsubame 2.0 from Japan, rated at 1.192 PF and the Hopper from USA, that reaches an overall performance of 1.05 PF.
The most notable new entry to the Top500 being Tsubame 2.0, the new supercomputer from Tokyo Institute of Technology, a system that delivers petaflop-class performance while remaining extremely efficient, consuming just 1.340 megawatts, dramatically less power than any other system on the top five.
"Tsubame 2.0 is an impressive achievement, balancing performance and power to deliver the most energy efficient petaflop-class supercomputer ever built," said Bill Dally, chief scientist at NVIDIA.
"The path to exascale computing will be forged by groundbreaking systems like Tsubame 2.0," concluded Mr. Dally, who, given his position within NVIDIA, probably had a direct involvement in helping the researchers at the Tokyo Institute of Technology leverage the huge computational potential provided by the company's Tesla GPUs.
And the reason why NVIDIA is so actively involved in this field is that GPUs have quickly become the enabling technology behind the world's top supercomputers, since they contain hundreds of parallel processor cores capable of dividing up large computational workloads and processing them simultaneously, significantly increasing system performance.