6

Using CUDA-Accelerated Libraries

The c-mex configuration opens many possibilities for what we can do with our MATLAB programs. By using CUDA-accelerated libraries available off the shelf, we can even push the limits far beyond what MATLAB can do. Using other third-party libraries in c-mex is no different from using them in regular C/C++ programs. In this chapter, we discuss how to use CUDA Basic Linear Algebra Subroutines (CUBLAS) for MATLAB through c-mex, the CUDA FFT library (CUFFT) for MATLAB through c-mex, and Thrust, a C++ template library for CUDA based on the Standard Template Library (STL).

Keywords

Linear algebra; FFT; c-mex debugging; C++ template library

6.1 Chapter Objectives

The c-mex configuration opens many possibilities ...

Get Accelerating MATLAB with GPU Computing now with the O’Reilly learning platform.

O’Reilly members experience books, live events, courses curated by job role, and more from O’Reilly and nearly 200 top publishers.