December 2012
Intermediate to advanced
514 pages
13h 6m
English
15.1 OpenACC Versus CUDA C
15.2 Execution Model
15.3 Memory Model
15.4 Basic OpenACC Programs
15.5 Future Directions of OpenACC
15.6 Exercises
The OpenACC Application Programming Interface (API) provides a set of compiler directives, library routines, and environment variables that can be used to write data-parallel FORTRAN, C, and C++ programs that run on accelerator devices, including GPUs. It is an extension to the host language. The OpenACC specification was initially developed by the Portland Group (PGI), Cray Inc., and NVIDIA, with support from CAPS enterprise. This chapter presents an introduction to OpenACC to parallel programmers ...
Read now
Unlock full access