CREAM: Computing in ReRAM-Assisted Energy- and Area-Efficient SRAM for Reliable Neural Network Acceleration
- 11 May 2023
- journal article
- research article
- Published by Institute of Electrical and Electronics Engineers (IEEE) in IEEE Transactions on Circuits and Systems I: Regular Papers
- Vol. 70 (8) , 3198-3211
Abstract
SRAM-based computing-in-memory (CIM) has been widely explored to accelerate neural networks (NNs). However, it is challenging to store all weights of many modern NNs due to limited on-chip SRAM capacity. This bottleneck induces a large amount of off-chip DRAM accesses and impedes the improvement of performance and energy efficiency. This paper proposes a new approach of computing in resistive random-access memory (ReRAM)-assisted energy-and area-efficient SRAM (CREAM) for accelerating large-scale NNs while eliminating the DRAM access. The NN weights are all stored in high-density on-chip ReRAMs and restored to the proposed non-volatile SRAM (nvSRAM) CIM cells with array-level parallelism. Furthermore, to deal with the influence of ReRAM and CMOS variations, a novel layer-wise and bit-wise weight-configuration search algorithm is proposed by leveraging different sensitivity of each layer in NN models. A data-aware weight-mapping method is also presented to efficiently map NN models to ReRAMs in CREAM for high computation parallelism. The experiment results show 10.3 $\times$ weight storage density over the standard 6T SRAM array. Evaluations of ResNet-18 and VGG-9 on CIFAR-10/CIFAR-100 datasets show up to 3.47 $\times$ and 1.70 $\times$ energy efficiency over two baseline designs of SRAM-CIM and ReRAM-CIM, respectively, in addition to 15.6% higher accuracy than ReRAM-CIM under device variations.Keywords
Funding Information
- NSFC (62174110, U21B2030, 92264204, 61934005)
- National Key Research and Development Program of China (2021YFA0717400, 2019YFA0706100, 2018YFA0701500)
- Natural Science Foundation of Shanghai Municipality (23ZR1433200)
- BirenTech Research
This publication has 39 references indexed in Scilit:
- 40× Retention Improvement by Eliminating Resistance Relaxation with High Temperature Forming in 28 nm RRAM Chip2018
- A Variation-Tolerant In-Memory Machine Learning Classifier via On-Chip TrainingIEEE Journal of Solid-State Circuits, 2018
- A 4M Synapses integrated Analog ReRAM based 66.5 TOPS/W Neural-Network Processor with Cell Current Controlled Writing and Flexible Network Architecture2018
- Design and Analysis of a Neural Network Inference Engine Based on Adaptive Weight CompressionIEEE Transactions on Computer-Aided Design of Integrated Circuits and Systems, 2018
- Average 7T1R Nonvolatile SRAM With R/W Margin Enhanced for Low-Power ApplicationIEEE Transactions on Very Large Scale Integration (VLSI) Systems, 2017
- Design of Nonvolatile SRAM with Ferroelectric FETs for Energy-Efficient Backup and RestoreIEEE Transactions on Electron Devices, 2017
- TETRISACM SIGARCH Computer Architecture News, 2017
- Compact Modeling of RRAM Devices and Its Applications in 1T1R and 1S1R Array DesignIEEE Transactions on Electron Devices, 2015
- A 0.13 µm 8 Mb Logic-Based Cu$_{\rm x}$Si$_{\rm y}$O ReRAM With Self-Adaptive Operation for Yield Enhancement and Power ReductionIEEE Journal of Solid-State Circuits, 2013
- NVSim: A Circuit-Level Performance, Energy, and Area Model for Emerging Nonvolatile MemoryIEEE Transactions on Computer-Aided Design of Integrated Circuits and Systems, 2012