Moscow State University is upgrading a already challenging Lomonosov supercomputer with a single more Nvidia Tesla GPUs.

 

The tricked-out complement is approaching to integrate 1,554 Tesla X2070 cards with an matching series of quad-core CPUs, delivering 1.3 petaflops of rise opening and- effectively creation Lomonosov a series a single supercomputer in Russia.

According to Victor Sadovnichy of Moscow State University, a complement is now use-d for computationally complete research, such as tellurian meridian change, sea modeling, post-genomic disinfectant and- universe arrangement simulations.

“Our investigate requires huge computational resources, and- you need to broach this opening as well as possible,” explained Sadovnichy.

“The usually approach for us to grasp these identical tiwn goals is with a hybrid GPU/CPU formed system.”

An Nvidia repute told TG Daily which a series of supercomputing centers in Russia have been seeking during ways of augmenting opening but surpassing energy budgets.

For example, Lobachevsky State University of Nizhni Novgorod (NNSU) – Russia’s initial CUDA Research Center – actively use-s GPUs opposite projects as different as study vital systems around plain mathematical displaying and- large-scale computation.

NNSU is additionally slated to implement a GPU-enabled cluster with a rise opening of 100 teraflops after this year, and- skeleton to enlarge a cluse-r to 500 teraflops by a finish of 2012.