- December 2007


More Chip Stories to Watch
The Channel Wire
"Is 2008 the year of accelerators? Examples include FPGA processors, Cell processors and Nvidia's Tesla. The industry has been sniffing around this issue for years. Intel's QuickAssist and AMD's Torrenza initiatives are vying for the industry standard."

Nvidia opens developer site
"The CUDA Zone site is geared to developers using CUDA (Compute Unified Device Architecture). CUDA is a C language environment for CUDA-enabled GPUs. The environment has helped engineers, scientists and others find ways to process the vast amounts of data required by their work and boost speed, Nvidia said."

Playing or processing?
The Economist Technology Quarterly
"The idea of using graphics chips for more general-purpose computing has been around for years, but only recently have the chips become sophisticated enough for it to work in practice. Both Nvidia and AMD, the world's largest graphics-card manufacturers, have decided that GPU computing could be a big opportunity. Nvidia has released a product line designed specifically for non-graphics applications, and a specialised programming language for use with it."

NVIDIA finds new markets
Euro Finanzen
"Euro Finanzen interviews Mike Hara following NVIDIA’s impressive financial results. NVIDIA sees a great growth opportunity in the field of HPC. Whenever a huge amount of data needs to be computed, a GPU is the better choice for the job."

Pixels to PetaFLOPS
Today’s HPC Clusters
"CUDA is very interesting because NVIDIA created a new model for programming general purpose computations on GPUs. CUDA is data parallel computing, using thousands of threads with a Parallel Data Cache to help increase arithmetic intensity for large performance boosts (arithmetic intensity refers to the compute intensity of the code). With CUDA, you can program in C and then use extensions to program for the GPUs. This feature allows you to target certain portions of the code for execution on the GPU and the rest to run on the CPU."