Massive video-surveillance parallelization on the Cell Broadband Engine processor

被引:1
作者
Rabie, T. [1 ]
Kidwai, H. K. [2 ]
Sibai, F. N. [2 ]
机构
[1] United Arab Emirates Univ, Fac Informat Technol, Al Ain, U Arab Emirates
[2] United Arab Emirates Univ, Coll Informat Technol, Al Ain, U Arab Emirates
关键词
Cell broadband engines; Parallelization; Synergistic processor elements;
D O I
10.1147/JRD.2010.2074930
中图分类号
TP3 [计算技术、计算机技术];
学科分类号
0812 ;
摘要
The Cell Broadband Engine (R) (Cell/B.E.(R)) is a multicore processor with a PowerPC (R) processor element (PPE) and eight synergistic processor elements (SPEs). The Cell/B.E. architecture is designed to improve upon conventional processors in terms of memory latency, bandwidth, and power consumption. In this paper, we discuss the parallelization, implementation, and performance of a video-surveillance application on the Cell/B.E. We report the video-surveillance application performance measured on a computer with one Cell/B.E. processor and with varying numbers of SPEs enabled. These results were compared with the results obtained on the single PPE of the Cell/B.E. with all eight SPEs disabled. The results indicate that our video-surveillance application performs approximately 25 times faster on the eight-SPE-core Cell/B.E. than modern reduced instruction set computer processors by processing input data from eight separate surveillance-video streams in parallel.
引用
收藏
页数:8
相关论文
共 19 条
[1]  
[Anonymous], 2007, CELL BROADB ENG PROG
[2]  
[Anonymous], 1982, Computer Vision
[3]  
[Anonymous], 2004, H.264 and MPEG-4 Video Compression: Video Coding for Next-generation Multimedia
[4]  
Arevalo A., 2008, Programming the Cell Broadband Engine Architecture Examples and Best Practices
[5]  
Burt P. J., 1989, Proceedings. Workshop on Visual Motion (IEEE Cat. No.89CH2716-9), P2, DOI 10.1109/WVM.1989.47088
[6]   OPTIC FLOW AND AUTONOMOUS NAVIGATION [J].
CAMPANI, M ;
GIACHETTI, A ;
TORRE, V .
PERCEPTION, 1995, 24 (03) :253-267
[7]  
CHEN T., 2005, CELL BROADBAND ENGIN
[8]  
Collins R.T., 2000, VSAM Final Rep, P1
[9]  
Gibson J. J., 2014, ECOLOGICAL APPROACH, DOI [DOI 10.4324/9781315740218, 10.4324/9781315740218]
[10]  
Irani M., 1992, Proceedings. 1992 IEEE Computer Society Conference on Computer Vision and Pattern Recognition (Cat. No.92CH3168-2), P216, DOI 10.1109/CVPR.1992.223272