NVIDIA Completes Fast Tesla Card, Breaks AMBER 11 Record

One would have definitely heard about how today was the day when NVIDIA was going to release a new video card, but it looks like the GeForce GTX 560 isn't the only adapter to grace the world with its presence, as the Tesla series also saw some movement.

NVIDIA may be best known for all its consumer graphics products and, more recently, the Tegra 2 mobile platform, but it definitely hasn't limited itself to just these areas.

In fact, seeing how supercomputers are granting more and more favor to parallel processing, GPU computing modules are one of the company's outlets as well.

More specifically, NVIDIA has the so-called Tesla series of cards which has just received its newest and fastest member yet.

Indeed, the so-called Tesla M2090 is not just the fastest of its line, but in the whole world, apparently.

Basically, NVIDIA made a point of mentioning, in its press announcement, that the board has broken the record for the highest performance ever returned under the AMBER 11 applications.

For those that don't know, AMBER 11 is a program that simulates biomolecule behavior. It was able to process 69 nanoseconds of simulation per day on four Tesla M2090 GPUs coupled with four CPUs.

Hardware-wise, there are 512 CUDA cores, as well as 6 GB of VRAM and support for such GPU-accelerated programs as MATLAB, GADGET2, GPU-BLAST, AMBER, GROMACS, NAMD, etc.

"Clients running intensive data center applications require systems that can process massive amounts of complex data quickly and efficiently," said Glenn Keels, director of marketing, Hyperscale business unit, HP.

"The decade long collaboration between HP and NVIDIA has created one of the industry's fastest CPU to GPU configurations available, delivering clients the needed processing power and speed to handle the most complex scientific computations."

Having already been included in the ProLian SL390 G7 4U server from HP, the NVIDIA Tesla M2090 is described in detail here.