A Fully Bit-Flexible Computation in Memory Macro Using Multi-Functional Computing Bit Cell and Embedded Input Sparsity Sensing
- 9 January 2023
- journal article
- research article
- Published by Institute of Electrical and Electronics Engineers (IEEE) in IEEE Journal of Solid-State Circuits
- Vol. 58 (5) , 1487-1495
Abstract
Computation in memory (CIM) overcomes the von Neumann bottleneck by minimizing the communication overhead between memory and process elements. However, using conventional CIM architectures to realize multiply-accumulate operations (MACs) with flexible input and weight bit precision is extremely challenging. This article presents a fully bit-flexible CIM design with a compact area and high energy efficiency. The proposed CIM macro employs a novel multi-functional computing bit cell design by integrating the MAC and the A/D conversion to maximize efficiency and flexibility. Moreover, an embedded input sparsity sensing and a self-adaptive dynamic range (DR) scaling scheme are proposed to minimize the energy-consuming A/D conversions in CIM. Finally, the proposed CIM macro implementation utilizes an interleaved placement structure to enhance the weight-updating bandwidth and the layout symmetry. The proposed CIM design fabricated in standard 28-nm CMOS technology achieves an area efficiency of 27.7 TOPS/mm $^2$ and an energy efficiency of 291 TOPS/W, demonstrating a highly energy-area-efficient flexible CIM solution.Keywords
Funding Information
- Ministry of Science and Technology, Taiwan (MOST 110-2218-E-002-034-MBK, MOST 111-2218-E-002-018-MBK)
- Intelligent and Sustainable Medical Electronics Research Fund in National Taiwan University
- MediaTek Inc (MTKC-2022-0125)
This publication has 27 references indexed in Scilit:
- 15.3 A 351TOPS/W and 372.4GOPS Compute-in-Memory SRAM Macro in 7nm FinFET CMOS for Machine-Learning Applications2020
- 15.2 A 28nm 64Kb Inference-Training Two-Way Transpose Multibit 6T SRAM Compute-in-Memory Macro for AI Edge Chips2020
- A Twin-8T SRAM Computation-in-Memory Unit-Macro for Multibit CNN-Based AI Edge ProcessorsIEEE Journal of Solid-State Circuits, 2019
- A 16K SRAM-Based Mixed-Signal In-Memory Computing Macro Featuring Voltage-Mode Accumulator and Row-by-Row ADC2019
- In-Memory Computing: Advances and ProspectsIEEE Solid-State Circuits Magazine, 2019
- A 64-Tile 2.4-Mb In-Memory-Computing CNN Accelerator Employing Charge-Domain ComputeIEEE Journal of Solid-State Circuits, 2019
- CONV-SRAM: An Energy-Efficient SRAM With In-Memory Dot-Product Computation for Low-Power Convolutional Neural NetworksIEEE Journal of Solid-State Circuits, 2018
- A Variation-Tolerant In-Memory Machine Learning Classifier via On-Chip TrainingIEEE Journal of Solid-State Circuits, 2018
- In-Memory Computation of a Machine-Learning Classifier in a Standard 6T SRAM ArrayIEEE Journal of Solid-State Circuits, 2017
- A 6-bit 600-MS/s 5.3-mW Asynchronous ADC in 0.13-$\mu{\hbox{m}}$ CMOSIEEE Journal of Solid-State Circuits, 2006