Accelerating Video-Mining Applications Using Many Small, General-Purpose Cores

Eric Li,Wenlong Li,Xiaofeng Tong,Jianguo Li,Yurong Chen,Tao Wang,Patricia P. Wang,Wei Hu,Yangzhou Du,Yimin Zhang,Yen-Kuang Chen

IEEE Micro（2008）

引用 13|浏览4

暂无评分

摘要

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.

查看译文

关键词

128-bit simd vectorization,real-time processing capability,many-core architecture,general-purpose core,key video-mining module,video-mining application,parallel speedup,general-purpose cores,video retrieval,video-mining applications,necessary performance goal,proposed architecture,indexation,thread level parallelism,indexing,multicore,real time processing,feature extraction,real time systems,data level parallelism,classification algorithms,image retrieval,simd,face,data mining

AI 理解论文

溯源树

样例

生成溯源树，研究论文发展脉络

Chat Paper

正在生成论文摘要