Cocco: Hardware-Mapping Co-Exploration towards Memory Capacity-Communication Optimization
Zhanhong Tan, Zijian Zhu, Kaisheng Ma · 2024
Memory is a critical design consideration in current data-intensive DNN accelerators, as it profoundly determines energy consumption, bandwidth requirements, and area costs. As DNN structures become more complex, a larger on-chip memory capacity is required to reduce data movement overhead, but at the expense of silicon costs. Some previous works have proposed memory-oriented optimizations, such as different data reuse and layer fusion schemes. However, these methods are not general and potent enough to cope with various graph structures.