BiRD: Bi-Directional Input Reuse Dataflow for Enhancing Depthwise Convolution Performance on Systolic Arrays

Depthwise convolution (DWConv) is an effective technique for reducing the size and computational requirements of convolutional neural networks. However, DWConv's input reuse pattern is not easily transformed into dense matrix multiplications, leading to low utilization of processing elements (PEs) on existing systolic arrays. In this paper, we introduce a novel systolic array dataflow mechanism called <i>BiRD</i>, designed to maximize input reuse and boost DWConv performance. BiRD utilizes two directions of input reuse and necessitates only minor modifications to a typical weight-stationary type systolic array. We evaluate BiRD on the Gemmini platform, comparing it with existing dataflow types. The results demonstrate that BiRD achieves significant performance improvements in computation time reduction, while incurring minimal area overhead and improved energy consumption compared to other dataflow types. For example, on a 32<inline-formula><tex-math notation="LaTeX">$\times{}$</tex-math></inline-formula>32 systolic array, it results in a 9.8% area overhead, significantly smaller than other dataflow types for DWConv. Compared to matrix multiplication-based DWConv, BiRD achieves a 4.7<inline-formula><tex-math notation="LaTeX">$\times{}$</tex-math></inline-formula> performance improvement for DWConv layers of MobileNet-V2, resulting in a 55.8% reduction in total inference computation time and a 44.9% reduction in energy consumption. Our results highlight the effectiveness of BiRD in enhancing the performance of DWConv on systolic arrays.