Journal of Computers, Vol 7, No 7 (2012), 1639-1646, Jul 2012
doi:10.4304/jcp.7.7.1639-1646
Parallel Implementation of Xvid Decoder on Multi-Core
Ying Liu, Fuxiang Gao, Shiyuan Wang
Abstract
With rapid development of multimedia technology, performance of computers is changing constantly. Today even the most ordinary computers have already equipped with multi-core processors. At the same time, high-quality videos have become main requirement of customers. Therefore it is a serious problem how to make video codec process a large number of video data efficiently on multi-core processors. It is a good way to develop the software which is compatible with multi-core. But most video codec has been already designed and developed for single-core processors. So it’s a good idea to transform the current sequential program into the parallel one by the parallelization runtime library. In this paper we choose Intel isomorphic quad-core processor as hardware platform, Linux as OS, and use Intel parallel runtime library TBB to transform the decoder. The transformation includes that Loop parallelization, memory parallelization, data parallelization, pipeline parallelization and task-level parallelization. Then, for testing, sequential program and parallel one run on the same environment respectively, and the final results show that after parallelization the performance has improved significantly
Keywords
Parallelization; TBB; Multi-core;Xvid Decoder
References
[1] L. Peng et al, “Memory Performance and Scalability of Intel‟s and AMD‟s Dual-Core Processors: A Case Study”, IEEE, 2007
[2] R. Merritt, “CPU Designers Debate Multi-core Future”, EETimes Online, February 2008, http://www.eetimes.com/showArticle.jhtml?articleID=206105179
[3] Lu Jinzheng, Zhou Dongmei. Visual C++ Audio/Video Processing Technology and Engineering Practice. Electronics Industry Press, 2009, 519-563.
[4] Zhou Shuxian. OpenMP-based Programming on Multi-core PC. SCIENCE & TECHNOLOGY INFORMATION, 2010, 9: 78-79.
[5] Xvid. http://www.xvid.org/, 2011
[6] Xie Xianghui, Hu Sutai, Li Hongliang. Multi-core/many-core processor and its influences on computer architecture design. Journal of Frontiers of Computer Science and Technology, 2008, 2(6), 641-650.
[7] TBB opensource. http://www.threadbuildingblocks.org, 2008.
[8] Wang Hai tao, Liu Shu fen. Parallel Computing Based on Linux Cluster. Computer Engineering, 2010, 1(36), 64-66.
[9] Xi Jie; Chen Jie. Parallel Intra Prediction on Multi-core Platform. Science Technology and Engineering, 2010, 10(6), 1379-1383.
[10] Video Sequence Download. http://bbs.chinavideo.org/viewthread.php?tid=1006, 2007.
Full Text:
PDF