A 98 GMACs/W 32-core vector processor in 65nm CMOS

This paper presents a high-performance dual-issue 32-core SIMD platform for image and video processing. Eight cores with a 4-ports L2 cache are connected by CIB bus as a cluster. Four clusters are connected by mesh network. The proposed hierarchical network can provide 192 GB/sintercore communication BW in average. To reduce coherence operation in large-scale SMP, an application specified protocol is proposed. Comparing with MOESI, 67.8% of L1 Cache energy can be saved in 32 cores case. It can achieve a peak performance of 375 GMACs and 98 GMACs/W in 65 nm CMOS.