A CDE Definition
(Single Instruction stream Multiple Data stream) The instruction execution architecture of a vector processor (a CPU or GPU that performs one operation on multiple sets of data simultaneously). It is typically used to add or multiply eight or more sets of numbers at the same time for multimedia encoding and rendering as well as scientific applications. Hardware registers are loaded with numbers, and the mathematical operation is performed on all registers simultaneously. Contrast with SISD and MIMD. See GPU.
(Graphics Processing Unit) A programmable logic chip (processor) specialized for display functions. The GPU renders images, animations and video for the computer's screen. GPUs are located on plug-in cards, in a chipset on the motherboard or in the same chip as the CPU (see diagram below). See logic chip.
A GPU performs parallel operations. Although it is used for 2D data as well as for zooming and panning the screen, a GPU is essential for smooth decoding and rendering of 3D animations and video. The more sophisticated the GPU, the higher the resolution and the faster and smoother the motion in games and movies. GPUs on stand-alone cards include their own memory (RAM), while GPUs in the chipset or CPU chip share main memory with the CPU.
Not Just Graphics Processing
Since GPUs perform parallel operations on multiple sets of data, they are increasingly used as vector processors for non-graphics applications that require repetitive computations. For example, in 2010, a Chinese supercomputer achieved the record for top speed using more than seven thousand GPUs in addition to its CPUs (see GPGPU). See graphics pipeline and multi-GPU.
Graphics Hardware Locations
An Integrated GPU
Before/After Your Search Term
Terms By Topic
Click any of the following categories for a list of fundamental terms.