NEC plans to start selling computing solutions powered by NVIDIA's Tesla GPUs.
One of the most recent trends in the HPC market is to enhance application performance by harnessing the parallel processing power of the GPU. NVIDIA views the HPC market as one of its major growth areas. NVIDIA's business model is to provide NVIDIA Tesla GPU computing solutions to end users through sales partners, and to allow the partners to build specialized systems.
The Global Scientific Information and Computing Center (GSIC) at the Tokyo Institute of Technology is already using a Tesla-powered supercomputer called "TSUBAME".
TSUBAME was upgraded in October 2008 with 170 NVIDIA Tesla S1070 GPU Systems, while the system continuously remained in operation. The newly upgraded system recorded 77.48 Teraflops in Linpack, ranking it favorably in the global Top 500 supercomputer listing.
The Tesla S1070 1U GPU system is based on the NVIDIA CUDA parallel architecture. This architecture is accessible through the C language programming environment that allows developers and researchers to easily tap into the parallel architecture of the GPU.