ACCELERATING VIDEO-MINING APPLICATIONS USING MANY SMALL, GENERAL-PURPOSE CORES

作者：Li Eric^*; Li Wenlong; Tong Xiaofeng; Li Jianguo; Chen Yurong; Wang Tao; Wang Patricia P; Hu Wei; Du Yangzhou; Zhang Yimin; Chen Yen Kuang

来源：IEEE Micro, 2008, 28(5): 8-21.

DOI：10.1109/MM.2008.64

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 19x TO 62x FROM 64 CORES AND GET AN EXTRA 2.3x SPEEDUP FROM 128-BIT SIMD VECTORIZATION ON THE PROPOSED ARCHITECTURE.

• 出版日期2008-10