A Cross-Layer Framework for Design Space Exploration and Variation Analysis of Compute-in-Memory DNN Accelerators using Non-Volatile Ferroelectric Capacitors
Author(s)
Read, James Loranger
Advisor(s)
Editor(s)
Collections
Supplementary to:
Permanent Link
Abstract
Using non-volatile “capacitive” crossbar arrays for compute-in-memory (CIM) offers higher energy and area efficiency compared to “resistive” crossbar arrays. However, the impact of device-to-device (D2D) variation and temporal noise on the system-level performance has not been explored yet. In this work, we provide an end-to-end methodology that incorporates experimentally measured D2D variation into the design space exploration from capacitive weight cell design, CIM array with peripheral circuits, to the inference accuracy of SwinV2-T vision transformer and ResNet-50 on the ImageNet dataset. Our framework further assesses the system's power, performance, and area (PPA) by considering cell design, circuit structure, and model selection. We explore the design space using an early stopping algorithm to produce optimal designs while meeting strict inference accuracy requirements. Overall findings suggest that the capacitive CIM system is robust against D2D variation and noise, outperforming its resistive counterpart by 6.95× and 14.1× for the optimal design in the figure of merit (TOPS/W×TOPS/mm2) for ResNet-50 and SwinV2-T respectively.
Sponsor
Date
2024-04-29
Extent
Resource Type
Text
Resource Subtype
Thesis