Skip to Main Content
Programming Massively Parallel Processors, 3rd Edition
book

Programming Massively Parallel Processors, 3rd Edition

by David B. Kirk, Wen-mei W. Hwu
November 2016
Intermediate to advanced content levelIntermediate to advanced
576 pages
18h 22m
English
Morgan Kaufmann
Content preview from Programming Massively Parallel Processors, 3rd Edition
Chapter 4

Memory and data locality

Abstract

This chapter introduces the concepts of memory bound application. It uses matrix multiplication to illustrate opportunities for reducing the number of global memory accesses. It then introduces the tiling technique where barrier synchronization is used to coordinate the timing of executing threads for improved locality and reduced global memory accesses. The tiling techniques, however, involve additional complexities in boundary checks. The chapter uses matrix multiplication to illustrate the additional boundary checks needed for a tiled kernel to be applicable to arbitrary matrix sizes. The chapter concludes with an overview of how usage of shared memory and registers can affect the number of thread ...

Become an O’Reilly member and get unlimited access to this title plus top books and audiobooks from O’Reilly and nearly 200 top publishers, thousands of courses curated by job role, 150+ live events each month,
and much more.
Start your free trial

You might also like

Programming Massively Parallel Processors, 4th Edition

Programming Massively Parallel Processors, 4th Edition

Wen-mei W. Hwu, David B. Kirk, Izzat El Hajj
Engineering a Compiler, 2nd Edition

Engineering a Compiler, 2nd Edition

Keith D. Cooper, Linda Torczon
Algorithms, 4th Edition

Algorithms, 4th Edition

Robert Sedgewick, Kevin Wayne

Publisher Resources

ISBN: 9780128119877