Premium
TH‐C‐BRA‐01: An introduction to GPU Computing
Author(s) -
Sharp G
Publication year - 2010
Publication title -
medical physics
Language(s) - English
Resource type - Journals
SCImago Journal Rank - 1.473
H-Index - 180
eISSN - 2473-4209
pISSN - 0094-2405
DOI - 10.1118/1.3469480
Subject(s) - computer science , cuda , graphics processing unit , general purpose computing on graphics processing units , graphics , software , coprocessor , parallel computing , range (aeronautics) , computational science , computer architecture , computer graphics (images) , programming language , materials science , composite material
The graphical processing unit (GPU) is a special‐purpose co‐processor used to perform fast calculation of 2D and 3D graphics commands for computer displays. Recently GPUs have begun to be used as general‐purpose vector processors for a wide range of scientific programs. GPU programming is used in domains such as medical imaging financial analysis and oil exploration. This talk will introduce the history and capabilities of modern GPU‐based computing systems. I will describe the hardware organization of GPUs and describe how multithreaded GPU programs map onto the hardware. Next I will introduce the CUDA language and software tools used to create GPU‐accelerated programs. Finally I will present an overview of existing libraries and applications. Learning Objectives: 1. Learn the history and capabilities of GPU‐based computing systems 2. Learn the hardware organization of GPUs 3. Learn tools and procedures for programming GPUs