Dell And Nvidia Back Affordable Supercomputing

Just days after launching a preconfigured graphics-based supercomputing cluster, Nvidia announced that it is partnering with Dell to offer workstations powered by Nvidia’s Tesla GPU technology.

Nvidia and Dell said May 7 that Dell is offering the Tesla C1060 GPU Computing Processor in its Precision R5400, T5500 and T7500 workstations, a move officials with both companies said creates personal supercomputing capabilities.

The Tesla C1060 chip is based on Nvidia’s CUDA massively parallel architecture, and is designed to bring supercomputing capabilities to users who normally would not have easy access to such computing power, officials said.

“The Dell Precision R5400, T7500 and T5500 together with the Tesla GPU computing processors is putting the power of supercomputing on the desktop,” Greg Weir, senior manager of Dell’s Product Group, said in a statement.

Nvidia has been bringing its graphics technology into the HPC (high-performance computing) and supercomputing spaces, where systems traditionally have been powered by CPUs. However, Nvidia officials argue that running systems with GPUs can offer the same computing power at a much lower cost.

That’s important in scientific and research environments, where users often have to fight for time and space on shared supercomputers, and buying such systems can cost millions of dollars, according to Nvidia and Dell officials.

Nvidia-powered workstations give users the power of a cluster computing system but at a fraction of the price, they said.