GPU computing is the use of a GPU (graphics processing unit) to do general purpose scientific and engineering computing.

The model for GPU computing is to use a CPU and GPU together in a heterogeneous computing model. The sequential part of the application runs on the CPU and the computationally-intensive part runs on the GPU. From the user’s perspective, the application just runs faster because it is using the high-performance of the GPU to boost performance.

The application developer has to modify their application to take the compute-intensive kernels and map them to the GPU. The rest of the application remains on the CPU. Mapping a function to the GPU involves rewriting the function to expose the parallelism in the function and adding “C” keywords to move data to and from the GPU.

GPU computing is enabled by the massively parallel architecture of NVIDIA’s GPUs called the CUDA architecture. The CUDA architecture consists of 100s of processor cores that operate together to crunch through the data set in the application.

The Tesla 10-series GPU is the second generation CUDA architecture with features optimized for scientific applications such as IEEE standard double precision floating point hardware support, local data caches in the form of shared memory dispersed throughout the GPU, coalesced memory accesses and so on.

Graphics chips started as fixed function graphics pipelines. Over the years, these graphics chips became increasingly programmable, which led NVIDIA to introduce the first GPU or Graphics Processing Unit. In the 1999-2000 timeframe, computer scientists in particular, along with researchers in fields such as medical imaging and electromagnetics started using GPUs for running general purpose computational applications. They found the excellent floating point performance in GPUs led to a huge performance boost for a range of scientific applications. This was the advent of the movement called GPGPU or General Purpose computing on GPUs.

The problem was that GPGPU required using graphics programming languages like OpenGL and Cg to program the GPU. Developers had to make their scientific applications look like graphics applications and map them into problems that drew triangles and polygons. This limited the accessibility of tremendous performance of GPUs for science.

NVIDIA realized the potential to bring this performance to the larger scientific community and decided to invest in modifying the GPU to make it fully programmable for scientific applications and added support for high-level languages like C and C++. This led to the CUDA architecture for the GPU.

The CUDA parallel hardware architecture is accompanied by the CUDA parallel programming model that provides a set of abstractions that enable expressing fine-grained and coarse-grain data and task parallelism. The programmer can choose to express the parallelism in high-level languages such as C, C++, Fortran or driver APIs such as OpenCL™ and DirectX™-11 Compute.

The first language support NVIDIA provided is for the C language. A set of C for CUDA software development tools enable the GPU to be programmed using C with a minimal set of keywords or extensions. Support for Fortran, OpenCL, et cetera will follow soon.

The CUDA parallel programming model guides programmers to partition the problem into coarse sub-problems that can be solved independently in parallel. Fine grain parallelism in the sub-problems is then expressed such that each sub-problem can be solved cooperatively in parallel.

The CUDA GPU architecture and the corresponding CUDA parallel computing model are now widely deployed with 100s of applications and nearly a 1000 published research papers. CUDA Zone lists many of these applications and papers.

NVIDIA Tesla C1060 GPU Computing Processor

The NVIDIA Tesla C1060 computing processor enables the transition to energy efficient parallel computing power by bringing the performance of a small cluster to a workstation. With 240 processor cores and a standard C compiler that simplifies application development, Tesla scales to solve the world's most important computing challenges-more quickly and accurately.

• Massively -parallel many core architecture with 240 processor cores
• Widely accepted , easy to learn CUDA C programing environment
• IEEE 754 single & double precision floating point units
• Asynchronous transfer capability
• Scale to multiple GPUs and harness the performance of thousands of processor cores
• 4 GB GLOBAL MEMORY
• Shared Data Memory
• High Speed , PCI-Express Gen 2.0 Data Transfer

Supermicro SS6016GT-TF-TC2 GPU-Integrated Server for Tesla C1060 GPGPU Cards

The SS6016GT GPU Supercomputing Servers establish Supermicro as the true global IT hardware leader in server architecture, performance, and Green computing. Generating massively parallel processing power and unrivaled networking flexibility with two double-width GPUs, up to 5 expansion slots or with InfiniBand networking options, in a 1U form factor, the SS6016GT is performance and quality optimized for the most computationally-intensive applications. Supermicro’s unique server designs with Gold Level power supplies, energy-saving motherboards and enterprise class server management optimize cooling for even the most demanding applications, providing the perfect technology platform for these impressive GPU Supercomputing Servers.

• Dual Quad/Dual-Core Intel Xeon processors 5500 series
• Intel® 5520 chipset with QPI up to 6.4GT/s
• Up to 96GB of Reg. ECC DDR3 DIMM 1333/1066/800 MHz SDRAM
• 2x NVIDIA Tesla C1060 GPU Cards - PCI-E 2.0 x4 (in x16 slot - Low-Profile)
• 3x hot-swap 3.5" drive bays
• Matrox G200eW graphics controller
• 8x counter-rotating fans with optimal fan speed control
• IPMI 2.0 with virtual media over LAN and KVM-ove-LAN
• Dual LAN with Intel 82576 Gigabit Ethernet controller
• 1400W high-efficiency Gold Level (93%) Power Supply with PMbus