A 28-nm RRAM/SRAM Collaborative CIM Accelerator Supporting RRAM-Endurance-Latency Awareness for Edge Fine-Tuning
- 17 June 2025
- journal article
- Published by Institute of Electrical and Electronics Engineers (IEEE) in IEEE Journal of Solid-State Circuits
- Vol. 60 (10) , 3626-3638
Abstract
The resistive random access memory (RRAM)-based computing-in-memory features high density and high energy efficiency on edge. However, fine-tuning RRAM-based SoCs remains challenging due to the inherent limitations of non-volatile memory (NVM) characteristics. This work proposes an NVM-endurance/latency-aware collaborative RRAM/static random access memory (SRAM) compute-in-memory (CIM) accelerator that addresses the difficulties associated with RRAM write operations during fine-tuning tasks for both CNN and transformer models. The major contributions are: 1) RRAM-most significant bit (MSB)–SRAM-least significant bit (LSB) (RMSL)-based collaborative CIM macros, mitigating RRAM cell flipping times and alleviating the endurance concern; 2) an RRAM-sparse-SRAM-dense (RSSD) weight updating engine, minimizing the long reading and writing latency associated with RRAM access; and 3) a row-wise pipeline weight gradient (WG) computing data flow with low-hardware overhead. With a maximum bit update of ten times per fine-tuning (20 epochs, from start to convergence), the system achieves an energy efficiency of 76.25 TOPS/W for the CIM macro and 22.07 TOPS/W for the overall system. Thanks to the bit-level collaborative CIM, the RRAM CIM macro achieves the same hardware utilization during fine-tuning and inference processes to support RRAM’s energy-efficient computing. The proposed CIM accelerator, fabricated using 28-nm CMOS technology, achieves up to $143{\times }$ RRAM endurance improvement, $117{\times }$ and $144{\times }$ reduction in RRAM write power and latency.Keywords
Funding Information
- National Natural Science Foundation of China (62495101)
- Shanghai 2024 “Science and Technology Innovation Action Plan” Fundamental Research Program in Integrated Circuit (24JD1400300)
- Peng Cheng Laboratory and China Mobile (PCL-CMCC) Foundation for Science and Innovation (2024ZY2B0070)
This publication has 37 references indexed in Scilit:
- Parameter-efficient fine-tuning of large-scale pre-trained language modelsNature Machine Intelligence, 2023
- A 55nm 32Mb Digital Flash CIM Using Compressed LUT Multiplier and Low Power WL Voltage Trimming Scheme for AI Edge Inference2022
- A compute-in-memory chip based on resistive random-access memoryNature, 2022
- A computing-in-memory macro based on three-dimensional resistive random-access memoryNature Electronics, 2022
- A four-megabit compute-in-memory macro with eight-bit precision based on CMOS and resistive random-access memory for AI edge devicesNature Electronics, 2021
- Softermax: Hardware/Software Co-Design of an Efficient Softmax for Transformers2021
- Tokens-to-Token ViT: Training Vision Transformers from Scratch on ImageNet2021
- One-transistor one-resistor (1T1R) cell for large-area electronicsApplied Physics Letters, 2018
- In-memory computing with resistive switching devicesNature Electronics, 2018
- Integrated HfO2-RRAM to achieve highly reliable, greener, faster, cost-effective, and scaled devices2017