Implementation of multi thread management system on mobile GPGPU

Author

Jangseo Ku ; Jong-Il Park ; Kwangyeob Lee ; Jae-Chang Kwak ; Nakwoong Eum ; Chanho Lee

Author_Institution

Dept. of Comput. Eng., Seokyeong Univ., Seoul, South Korea

fYear

2012

fDate

19-22 Nov. 2012

Firstpage

1

Lastpage

4

Abstract

This thesis is intended to design multi thread mobile GPGPU optimized in mobile environment, and to verify an effective thread management method of the multi thread mobile processor. For the verification of the multi thread management method, Lane detection algorithm was implemented to compare nVidia´s CUDA Architecture and the designed GPGPU in terms of thread management efficiency. The number of thread is normalized to 48 threads. An implemented Lane Detection Algorithm is composed of Gaussian filter algorithm and Sobel Edge Detection algorithm. As a result, the designed GPGPU´s thread efficiency is up to 2 times higher than CUDA´s thread efficiency.

Keywords

edge detection; graphics processing units; mobile computing; multi-threading; object detection; parallel architectures; traffic engineering computing; Gaussian filter algorithm; Sobel edge detection algorithm; lane detection algorithm; mobile GPGPU; mobile environment; multithread management system; multithread mobile processor; nVidia CUDA architecture; Detection algorithms; Educational institutions; Filtering algorithms; Graphics processing units; Instruction sets; Mobile communication; Synchronization; GPGPU; Multi Thread; Parallel Processing; Thread Management;

fLanguage

English

Publisher

ieee

Conference_Titel

TENCON 2012 - 2012 IEEE Region 10 Conference

Conference_Location

Cebu

ISSN

2159-3442

Print_ISBN

978-1-4673-4823-2

Electronic_ISBN

2159-3442

Type

conf

DOI

10.1109/TENCON.2012.6412191

Filename

6412191