NVIDIA Tesla Used in Russian Academy of Sciences Supercomputer

NVIDIA's Tesla line of GPU compute accelerators are definitely showing that supercomputing likes parallel processing, this time through the actions of the Russian Academy of Sciences (RAS).

Or perhaps it would be more accurate to say that RAS is showing, through use of Tesla M2090, that yes, GPUs are favored in supercomputing.

Russian Academy of Sciences is a broad network of research institutes that look into hydro-dynamics, geological modeling, genomic analysis, computed tomography, electromagnetics, etc.

What they did now was install 128 Tesla M2090 GPUs in HP ProLiant SL390 G7 servers.

This mirrors, in a way, how the number of applications that can use GPU acceleration has grown over the past couple of years.

“GPU computing has been instrumental in the development of program and algorithmic simulations and borehole geophysical research,” said Vyacheslav Glinskikh, Ph.D. (Geophysics), head of Laboratory at Trofimuk Institute of Petroleum Geology and Geophysics SB RAS.

“Based on our results, we were able to create new automated geophysical data interpretation systems for the oil and gas industry, promising to dramatically improve efficiencies in oil and gas exploration.”

Right now, among those using GPU computing are Institute for Mathematics and Mechanics of UB RAS, Keldysh Institute of Applied Mathematics RAS, Institute for Cytology and Genetics of SB RAS, Institute for Image Processing Systems and Siberian Supercomputing Center based on Institute of Computational Mathematics and Mathematical Geophysics SB RAS.

The list, as some may have guessed, goes on, especially with AMD also offering GPU compute accelerators of its own.

“In my research using industrial codes for free surface flows using Navier-Stokes and shallow water equations, I was able to process research data much more efficiently using GPUs,” said Evstigneev Nikolay, senior staff scientist at the Laboratory of Chaotic and Nonlinear Dynamics, Institute for System Analysis RAS.

“This enabled me to analyze and monitor five times more dam break scenarios and flood regions.”