IBM Installs Nvidia Tesla Computing Accelerators into HPC Servers

Nvidia Tesla M2050 Now Available in IBM System X iDataPlex dx360 M3

by Anton Shilov
05/18/2010 | 12:46 PM

Nvidia Corp. said on Tuesday that its high-performance M2050 modules based on Tesla T20-series computing processors are now available inside IBM System x-series iDataPlex dx360 M3 high-performance servers. This is the first time when a major maker of servers integrates Nvidia Tesla computing solution into its high-performance computing (HPC) offering.

 

IBM System x iDataPlex dx360 M3 is system that comes in a half-depth 2U Flex or 3U storage chassis that is based on up to two Intel Xeon 5600-series (six-core, 3.06GHz) that can sport up to 128GB DDR3 memory, up to 24TB of storage space with RAID support and up to two Nvidia Tesla M2050 computing modules. Other specifications and configurations may wary depending on the actual requirements.

“Nvidia provides an innovative solution for customers who push the envelope in high-performance computing. GPU acceleration provides performance boosts for many applications in energy exploration, science and financial services. It is among the significant emerging supercomputer technologies to watch in the years ahead,” said Dave Turek, vice president of deep computing at IBM.

Nvidia Tesla M2050 and C2050 computing solutions based on the Fermi-generation graphics processors designed for high-performance computing features 448 stream processing units and carry 3GB of GDDR5 memory. Pricing of the M2050 and C205 is around $2500. The HPC products are used for financial modeling, oil and gas, federal and defense as well as key research codes in bioscience, such as AMBER, GROMACS and NAMD.

It is a major win for Nvidia to integrate Tesla computing accelerators into an HPC server by a major maker. The move not only enables high-profile customers to take advantage of GPGPU technology, but also greatly adds credibility to Nvidia’s Tesla computing solutions.