Browse > Article
http://dx.doi.org/10.7471/ikeee.2012.16.4.310

Method of Multi Thread Management based on Shader Instruction for Mobile GPGPU  

Lee, Kwang-Yeob (Dept. of Computer Engineering, Seokyeong University)
Park, Tae-Ryong (Dept. of Computer Engineering, Seokyeong University)
Publication Information
Journal of IKEEE / v.16, no.4, 2012 , pp. 310-315 More about this Journal
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. In thread management, there is no management hardware and implement with software instructions. 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 Land 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
GPGPU; Multi Thread; Thread Management; Parallel Processing;
Citations & Related Records
연도 인용수 순위
  • Reference
1 HyungKi Jeong, "A Design of a Multi-Threaded & Multi-Core GP-GPU using a dynamic Thread management techniques", The Graduate School of Seokyeong Universiry, August 2010.
2 NVIDIA, "NVIDIA CUDA C Programming Guide", November 2011.
3 NVIDIA, "CUDA API REFERENCE MANUAL", January 2012.
4 Hongfei Yu, Wei Liu, Jianghua Pu, Bobo Duan, Huai Yuan, Hong Zhao, "Lane recognition based on location of raised pavement markers", Intelligent Vehicles Symposium (IV), July 2011.
5 Shih-Shinh Huang, Chung-Jen Chen, Pei-Yung Hsiao, and Li-Chen Fu, "On-Board Vision System for Lane Recognition and Front-Vehicle Detection to Enhance Driver's Awareness", IEEE Intermtbnal confennu on Robdm a Automlltlon, April 2004.