The Community for Technology Leaders
Green Image
Issue No. 05 - September/October (2008 vol. 28)
ISSN: 0272-1732
pp: 8-21
Wei Hu , Intel
Wenlong Li , Intel
Yurong Chen , Intel
Tao Wang , Intel
Yimin Zhang , Intel
Eric Li , Intel
Yangzhou Du , Intel
Jianguo Li , Intel
ABSTRACT
Emerging video-mining applications such as image and video retrieval and indexing will require real-time processing capabilities. A many-core architecture with 64 small, in-order, general-purpose cores as the accelerator can help meet the necessary performance goals and requirements. The key video-mining modules can achieve parallel speedups of 19 to 62 from 64 cores and get an extra 2.3 speedup from 128-bit SIMD vectorization on the proposed architecture.
INDEX TERMS
video mining, accelerator, multicore, SIMD, thread-level parallelism, data-level parallelism
CITATION
Wei Hu, Yen-Kuang Chen, Wenlong Li, Yurong Chen, Tao Wang, Yimin Zhang, Eric Li, Xiaofeng Tong, Patricia P. Wang, Yangzhou Du, Jianguo Li, "Accelerating Video-Mining Applications Using Many Small, General-Purpose Cores", IEEE Micro, vol. 28, no. , pp. 8-21, September/October 2008, doi:10.1109/MM.2008.64
92 ms
(Ver )