Litcius/Paper detail

A CFMB STT-MRAM-Based Computing-in-Memory Proposal With Cascade Computing Unit for Edge AI Devices

Yongliang Zhou, Zixuan Zhou, Yiming Wei, Zhen Yang, Xiao Lin, Chenghu Dai, Licai Hao, Chunyu Peng, Hao Cai, Xiulong Wu

2023IEEE Transactions on Circuits and Systems I Regular Papers16 citationsDOI

Abstract

The application of non-volatile memory technology is increasingly attractive for Computing-in-memory (CIM) owing to high integration density and negligible standby power consumption. This study proposes an spin-transfer-torque (STT) magnetic random access memory (MRAM) based CIM macro which incorporates following innovative features: 1) cross-feedback margin-boost (CFMB) scheme to enable robust and fast reading operations against process variation and limited Tunneling Magnetoresistance Ratio (TMR); 2) cascade computing units (CCU) and related design method for efficient and stable multi-bit multiply-and-accumulate (MAC) operation; and 3) dual computing mode scheme and resolution adjustable quantization module to optimize energy efficiency and operating speed. The post-simulations are performed under 28nm CMOS&MTJ technology. The results demonstrate the achievement in energy efficiency of 36.4 TOPS/W while performing MAC operations with up to 16-bit weights, 4-bit inputs, and 22-bit outputs.

Topics & Concepts

Magnetoresistive random-access memoryComputer scienceCMOSCascadeTunnel magnetoresistanceEfficient energy useEnergy consumptionComputer hardwareParallel computingElectronic engineeringElectrical engineeringEngineeringRandom access memoryFerromagnetismQuantum mechanicsChemical engineeringPhysicsFerroelectric and Negative Capacitance DevicesMagnetic properties of thin filmsAdvanced Memory and Neural Computing
A CFMB STT-MRAM-Based Computing-in-Memory Proposal With Cascade Computing Unit for Edge AI Devices | Litcius