FPSpike: A Fully Parallel and Reconfigurable Architecture for Accelerating Spiking Neural Networks With Structured Sparsity
FPSpike: A Fully Parallel and Reconfigurable Architecture for Accelerating Spiking Neural Networks With Structured Sparsity
Abstract
This paper introduces FPSpike, a fully-parallel and reconfigurable architecture for accelerating spiking neural networks (SNNs) with structured sparsity. By introducing structured sparse synaptic connections, the neuron computation and weight storage costs are significantly reduced, while the wiring constraints of hardware implementation are alleviated, thus realizing a fully parallel SNN architecture on a single chip. Furthermore, FPSpike achieves high reconfigurability through local connections, allowing the network topology to be flexibly partitioned into multiple distinct hardware cores without redundancy, thereby enabling multi-task spatial parallel processing. In particular, a series of software-hardware co-designs are performed to improve the performance of FPSpike. Experimental results with FPGA-based implementation demonstrate that FPSpike achieves inference throughput improvements of 2.5× ∼ 43.9× and 3.2× ∼ 93.8× compared to Intel i7-12700K CPU and NVIDIA RTX3060Ti GPU, respectively. FPSpike also achieves performance speedups of 2.0× ∼ 20.2× and energy efficiency improvements of 1.3× ∼ 7.7× compared to the stateof-the-art FPGA-based SNN accelerators.
Thank you for visiting this project page. Register this project to explore its novelty and implementation support.
