Main Content

GPU Algorithm Acceleration

Accelerate your code using basic GPU computing

To speed up your code, you can try using your computer’s GPU. If all the functions that you want to use are supported on the GPU, you can simply use the gpuArray function to transfer input data to the GPU, and call the gather function to retrieve the output data from the GPU. For deep learning, MATLAB® provides automatic parallel support for multiple GPUs. You need Parallel Computing Toolbox™ to enable GPU support.

For a list of functions that accept GPU arrays, see Function List (GPU Arrays).


gatherTransfer distributed array or gpuArray to local workspace
gpuArrayArray stored on GPU


Run MATLAB Functions on a GPU (Parallel Computing Toolbox)

Hundreds of functions in MATLAB and other toolboxes run automatically on a GPU if you supply a gpuArray (Parallel Computing Toolbox) argument.

GPU Support by Release (Parallel Computing Toolbox)

Support for NVIDIA® GPU architectures by MATLAB release.

Run MATLAB Functions on Multiple GPUs (Parallel Computing Toolbox)

This example shows how to run MATLAB code on multiple GPUs in parallel, first on your local machine, then scaling up to a cluster.

Deep Learning with MATLAB on Multiple GPUs (Deep Learning Toolbox)

Speed up deep neural network training using multiple GPUs locally or in the cloud.

Pedestrian and Bicyclist Classification Using Deep Learning (Radar Toolbox)

Classify pedestrians and bicyclists based on their micro-Doppler characteristics using a deep learning network and time-frequency analysis.

GPU Acceleration of Scalograms for Deep Learning (Wavelet Toolbox)

Use your GPU to accelerate feature extraction for signal classification.

Related Information