← 返回 JSSC 论文列表
📄 下载 JSSC 原文 PDF
JSSC 2021第1期Memory7nmSRAMNeural Network Accelerator

A 7-nm Compute-in-Memory SRAM Macro Supporting Multi-Bit Input, Weight and Output and Achieving 351 TOPS/W and 372.4 GOPS Mahmut E. Sinangil , Member , IEEE

7纳米工艺下支持多比特输入权重和输出的存内计算SRAM宏设计
0.8V电源电压,5.5ns访问时间,351 TOPS/W能效,372.4 GOPS吞吐量
存内计算SRAM宏7纳米工艺FinFET乘加运算
采用7纳米FinFET工艺的标准双端口编译器宏构建存内计算(CIM)宏
通过读字线(RWL)脉冲数量表示4位输入,电荷共享实现4位权重
利用4位Flash ADC中感测放大器(SA)固有电容形成计算电容单元,节省面积并减少回踢效应
Abstract
In this work, we present a compute-in-memory (CIM) macro built around a standard two-port compiler macro using foundry 8T bit-cell in 7-nm FinFET technology. The pro- posed design supports 1024 4 b × 4 b multiply-and-accumulate (MAC) computations simultaneously. The 4-bit input is repre- sented by the number of read word-line (RWL) pulses, while the 4-bit weight is realized by charge sharing among binary-weighted computation caps. Each unit of computation cap is formed by the inherent cap of the sense amplifier (SA) inside the 4-bit Flash ADC, which saves area and minimizes kick-back effect. Access time is 5.5 ns with 0.8-V power supply at room temperature. The proposed design achieves energy efficiency of 351 TOPS/W and throughput of 372.4 GOPS. Implications of our design from neural network implementation and accuracy perspectives are also discussed.