GPU Computing
This feature is not supported on the Wolfram Cloud.

With the Wolfram Language, the enormous parallel processing power of Graphical Processing Units (GPUs) can be used from an integrated built-in interface. Incorporating GPU technology into the Wolfram Language allows high-performance solutions to be developed in many areas such as financial simulation, image processing, and modeling. GPU program creation and deployment is fully integrated with the Wolfram Language's high-level development tools and this gives a productivity boost to move from prototype to large-scale solution.


GPU Computing Using CUDALink »

CUDAInformation lists all CUDA device information

CUDAImageConvolve convolves images with specified kernel

CUDAFunctionLoad loads a user-defined function to run on a GPU using CUDA

CUDAErosion  ▪  CUDADilation  ▪  CUDAFourier  ▪  CUDADot  ▪  ...

CUDAMemoryLoad  ▪  CUDAMemoryAllocate  ▪  SymbolicCUDAFunction  ▪  ...

GPU Computing Using OpenCLLink »

OpenCLInformation lists all OpenCL device information

OpenCLFunctionLoad loads a user-defined function to run on a GPU using OpenCL

OpenCLMemoryLoad  ▪  OpenCLMemoryAllocate  ▪  SymbolicOpenCLFunction  ▪  ...