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 Programming 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 Programming Using OpenCLLink »

OpenCLInformation — lists all OpenCL device information

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

OpenCLMemoryLoad   ▪ OpenCLMemoryAllocate   ▪ SymbolicOpenCLFunction   ▪ ...

GPU Computing »

GPUArray — array stored in memory accessible for GPU computation

LinearSolve   ▪ LeastSquares   ▪ Fourier   ▪ RandomReal   ▪ Mean   ▪ Sort   ▪ ...

Related Tech Notes

▪
• CUDALink User Guide
▪
• OpenCLLink User Guide

Related Guides

▪
• CUDA Computation in the Wolfram Language
▪
• OpenCL Computation in the Wolfram Language
▪
• GPU Computing in the Wolfram Language
▪
• Parallel Computing in the Wolfram Language
▪
• Calling External Programs
▪
• Summary of New Features in 8.0