Archive notice: This article was originally published on May 7, 2009. Links and embedded videos are preserved as part of the historical record.
Research stood to benefit particularly from mass-market availability of NVIDIA high-performance computing. Researchers at institutions including Harvard, Cambridge and the Tokyo Institute of Technology competed for time on shared supercomputers consuming hundreds of kilowatts and costing millions of euros to implement and maintain. Dell Precision workstations equipped with Tesla GPUs offered each researcher a “personal supercomputer” with cluster-class computing power at one hundredth of the price.
“By integrating Tesla GPUs, Dell Precision R5400, T5500 and T7500 workstations offer supercomputer performance on the desktop,” said Dell Product Group Senior Manager Greg Weir.
National Instruments LabVIEW Product Manager Jeff Meisel said the company was developing the control system for a European telescope project creating the world’s largest telescope. A CUDA interface for LabVIEW simulated and checked its 984-segment M1 mirror. One Dell workstation with a Tesla C1060 achieved near-real-time monitoring of the mirror simulation, which Meisel said would not have been possible without GPU computing power.
