Problem
Research questions and friction points this paper is trying to address.
Compute-in-memory
spiking neural networks
flexibility
workloads
FlexSpIM: An Event-Based Digital Compute-In-Memory Accelerator with Flexible Operand Resolution and Layer-Wise Hybrid Stationarity
🤖 AI Summary
本文提出FlexSpIM，一种支持任意操作数分辨率和形状的数字计算内存架构，通过混合数据流减少能耗和延迟，提高SNN在边缘视觉应用中的效率。
📝 Abstract
Compute-in-memory (CIM) accelerators for spiking neural networks (SNNs) offer a promising solution for achieving $\mu$s-level inference latency and ultra-low energy in edge vision applications. However, their limited flexibility at both circuit and system levels restricts their deployment across diverse workloads. This work introduces FlexSpIM, a digital CIM architecture supporting arbitrary operand resolution and shape within a unified storage for weights and neuron states (i.e., membrane potentials). These circuit-level capabilities enable a layer-level hybrid weight- and output-stationary dataflow, maximizing operand reuse and reducing costly on- and off-chip data movement during SNN execution. Measurement results from a fabricated FlexSpIM prototype in 40-nm CMOS demonstrate competitive 1-bit-normalized energy efficiency and higher throughput compared with prior fixed-precision digital CIM-based SNN accelerators, while providing bitwise resolution reconfiguration. Evaluated on the IBM DVS gesture dataset, FlexSpIM achieves 95.8% accuracy while enabling up to 45% energy and 52% latency reductions in large-scale systems compared with fixed stationarity approaches.
Innovation
Methods, ideas, or system contributions that make the work stand out.
Compute-in-memory
Flexible Operand Resolution
Hybrid Stationarity
SNN Accelerators
🔎 Similar Papers
No similar papers found.
