Chapter 2

Data parallel computing

David Luebke

Abstract

This chapter introduces the concept of data parallelism and the essential CUDA C feature for writing a simple CUDA C program. It starts with the concept of threads, host, and device. It introduces CUDA device memory management and data transfer API functions. It further introduces the basic structure of a CUDA C kernel function, built-in variables, function declaration keywords, and kernel launch syntax.

Keywords

Data parallelism; scalable parallel program; thread; kernel; API; RGB; greyscale; kernel launch; execution configuration n parameters; data transfer; error handling; stub function; SPMD

Get Programming Massively Parallel Processors, 3rd Edition 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.