Chapter 8. Using CUDA Libraries

In this penultimate chapter, we take a look at some of the software libraries that are available to provide you with access to CUDA capabilities without having to write all the code yourself. As we tour the libraries, we will explicitly identify some common parallel programming patterns and look at examples illustrating how the libraries make those patterns available for your use. Before we are done, you will see library-based alternative implementations of apps for image processing and computing distances, dot products, and centroids, along with new examples involving Monte Carlo estimation of pi (π), computing distance in image space, and solving systems of linear equations.

Custom versus Off-the-Shelf

Up to ...

Get CUDA for Engineers: An Introduction to High-Performance Parallel Computing now with O’Reilly online learning.

O’Reilly members experience live online training, plus books, videos, and digital content from 200+ publishers.