O'Reilly logo

Professional CUDA C Programming by Ty McKercher, Max Grossman, John Cheng

Stay ahead with the world's most comprehensive technology and business learning platform.

With Safari, you learn the way you learn best. Get unlimited access to videos, live online training, learning paths, books, tutorials, and more.

Start Free Trial

No credit card required

Chapter 8GPU-Accelerated CUDA Libraries and OpenACC

What's in this chapter?

  • Exploring new levels of parallelism with CUDA libraries
  • Understanding the common workflow shared by many CUDA libraries
  • Experimenting with CUDA libraries in linear algebra, Fourier transforms, and random number generation
  • Covering new library features in CUDA 6
  • Accelerating applications on GPUs with OpenACC directives

In this book, you have learned a wide range of basic and advanced features in CUDA C. These lessons have enabled you to take advantage of the computational throughput of GPUs when writing new, custom applications or porting existing, legacy applications by hand. However, in many cases the main barrier to building applications in CUDA is development time. It is imperative that you maximize productivity and efficiency when creating or porting applications.

To augment the abilities of CUDA developers, NVIDIA and other institutions provide domain-specific CUDA libraries that can be used as building blocks for more complex applications. These libraries have been optimized by CUDA experts and designed to have high-level, highly-usable APIs with standardized data formats to facilitate their ability to plug in to existing applications (pluggability). CUDA libraries ...

With Safari, you learn the way you learn best. Get unlimited access to videos, live online training, learning paths, books, interactive tutorials, and more.

Start Free Trial

No credit card required