A Review of Memory Wall for Neuromorphic Computing

TL;DR

This review analyzes various memory technologies used in FPGA-based neuromorphic computing, comparing their performance, limitations, and suitability for AI applications to guide future research and development.

Contribution

It provides a comprehensive comparison of memory types for neuromorphic FPGA systems, highlighting their strengths, limitations, and practical implementation insights.

Findings

SRAM offers low latency but limited scalability.

Emerging non-volatile memories like ReRAM show promise for energy efficiency.

High-bandwidth memory improves data transfer rates for neural networks.

Abstract

This paper reviews memory technologies used in Field-Programmable Gate Arrays (FPGAs) for neuromorphic computing, a brain-inspired approach transforming artificial intelligence with improved efficiency and performance. It focuses on the essential role of memory in FPGA-based neuromorphic systems, evaluating memory types such as Static Random-Access Memory (SRAM), Dynamic Random-Access Memory (DRAM), High-Bandwidth Memory (HBM), and emerging non-volatile memories like Resistive RAM (ReRAM) and Phase-Change Memory (PCM). These technologies are analyzed based on latency, bandwidth, power consumption, density, and scalability to assess their suitability for storing and processing neural network models and synaptic weights. The review provides a comparative analysis of their strengths and limitations, supported by case studies illustrating real-world implementations and performance outcomes. This review offers insights to guide researchers and practitioners in selecting and optimizing memory technologies, enhancing the performance and energy efficiency of FPGA-based neuromorphic platforms, and advancing applications in artificial intelligence.