MLFlash-CIM: Embedded Multi-Level NOR-Flash Cell based Computing in Memory Architecture for Edge AI Devices

被引：2

作者：

Zeng, Sitao ^{[1
]}

Zhang, Yuxin ^{[1
]}

Zhu, Zhiguo ^{[1
]}

Qin, Zhaolong ^{[1
]}

Dou, Chunmeng ^{[2
]}

Si, Xin ^{[1
,3
]}

Li, Qiang ^{[1
]}

机构：

[1] Univ Elect Sci & Technol China, Chengdu, Peoples R China

[2] Chinese Acad Sci, Inst Microelect, Beijing, Peoples R China

[3] Southeast Univ, Nanjing, Peoples R China

来源：

2021 IEEE 3RD INTERNATIONAL CONFERENCE ON ARTIFICIAL INTELLIGENCE CIRCUITS AND SYSTEMS (AICAS) | 2021年

关键词：

Multi-Level NOR-Flash; Computing in memory; Artificial Intelligence; Convolutional Neural Network;

D O I：

10.1109/AICAS51828.2021.9458438

中图分类号：

TP18 [人工智能理论];

学科分类号：

081104 ; 0812 ; 0835 ; 1405 ;

摘要：

Computing-in-Memory (CIM) is a promising method to overcome the well-known "Von Neumann Bottleneck" with computation insides memory, especially in edge artificial intelligence (AI) devices. In this paper, we proposed a 40nm 1Mb Multi-Level NOR-Flash cell based CIM (MLFlash-CIM) architecture with hardware and software co-design. Modeling of proposed MLFlash-CIM was analyzed with the consideration of cell variation, number of activated cells, integral non-linear (INL) and differential non-linear (DNL) of input driver, and quantization error of readout circuits. We also proposed a multi-bit neural network mapping method with 1/n top values and an adaptive quantization scheme to improve the inference accuracy. When applied to a modified VGG-16 Network with 16 layers, the proposed MLFlash-CIM can achieve 92.73% inference accuracy under CIFAR-10 dataset. This CIM structure also achieved a peak throughput of 3.277 TOPS and an energy efficiency of 35.6 TOPS/W for 4-bit multiplication and accumulation (MAC) operations.

引用

页数：4