PICO-RAM: A PVT-Insensitive Analog Compute-In-Memory SRAM Macro With In Situ Multi-Bit Charge Computing and 6T Thin-Cell-Compatible Layout

Analog compute-in-memory (CIM) in static random access memory (SRAM) is promising for accelerating deep learning inference by circumventing the memory wall and exploiting ultra-efficient analog low-precision arithmetic. Latest analog CIM designs attempt bit-parallel (BP) schemes for multi-bit analog matrix-vector multiplication (MVM), aiming at higher energy efficiency, throughput, and training simplicity and robustness over conventional bit-serial (BS) methods that digitally shift-and-add multiple partial analog computing results. However, BP operations require more complex analog computations and become more sensitive to well-known analog CIM challenges, including large cell areas, inefficient and inaccurate multi-bit analog operations, and vulnerability to PVT variations. This article presents PICO-RAM, a PVT-insensitive and compact CIM SRAM macro with charge-domain BP computation. It adopts a multi-bit thin-cell multiply-accumulate (MAC) unit that shares the same transistor layout as the most compact 6T SRAM cell. All analog computing modules, including digital-to-analog converters (DACs), MAC units, analog shift-and-add, and analog-to-digital converters (ADCs), reuse one set of local capacitors inside the array, performing in situ computation to save area and enhance accuracy. A compact 8.5-bit dual-threshold time-domain ADC power gates the main path most of the time, leading to a significant energy reduction. Our 65-nm prototype achieves the highest weight storage density of 559 Kb/mm2 and exceptional robustness to temperature and voltage variations (<inline-formula xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink"> <tex-math notation="LaTeX">${-} 40~^{\circ } $ </tex-math></inline-formula> C to <inline-formula xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink"> <tex-math notation="LaTeX">$105~{^{\circ }}$ </tex-math></inline-formula> C and 0.65 to 1.2 V) among SRAM-based analog CIM designs.

Authors

Institutions

Publication Details

Journal
IEEE Journal of Solid-State Circuits
Published
2024-08-13
DOI
https://doi.org/10.1109/jssc.2024.3422826
Citations
17
Primary Topic
Semiconductor materials and devices
Type
article
Field-Weighted Citation Impact
3.55
Controls
|||
ALL TIME
JAN
FEB
MAR
APR
MAY
JUN
JUL
AUG
SEP
OCT
article

PICO-RAM: A PVT-Insensitive Analog Compute-In-Memory SRAM Macro With In Situ Multi-Bit Charge Computing and 6T Thin-Cell-Compatible Layout

Yubei Chen, Wei-Chen Wei, Akhil Pakala, Yiwei Zou et al.
17 citations
IEEE Journal of Solid-State Circuits
Semiconductor materials and devices
3.55
article

PICO-RAM: A PVT-Insensitive Analog Compute-In-Memory SRAM Macro With In Situ Multi-Bit Charge Computing and 6T Thin-Cell-Compatible Layout

Yubei Chen, Wei-Chen Wei, Akhil Pakala, Yiwei Zou, Zhiyu Chen, Kaiyuan Yang, Weier Wan, Zihao Li, Ziyuan Wen
article en
17 citations

Abstract

Analog compute-in-memory (CIM) in static random access memory (SRAM) is promising for accelerating deep learning inference by circumventing the memory wall and exploiting ultra-efficient analog low-precision arithmetic. Latest analog CIM designs attempt bit-parallel (BP) schemes for multi-bit analog matrix-vector multiplication (MVM), aiming at higher energy efficiency, throughput, and training simplicity and robustness over conventional bit-serial (BS) methods that digitally shift-and-add multiple partial analog computing results. However, BP operations require more complex analog computations and become more sensitive to well-known analog CIM challenges, including large cell areas, inefficient and inaccurate multi-bit analog operations, and vulnerability to PVT variations. This article presents PICO-RAM, a PVT-insensitive and compact CIM SRAM macro with charge-domain BP computation. It adopts a multi-bit thin-cell multiply-accumulate (MAC) unit that shares the same transistor layout as the most compact 6T SRAM cell. All analog computing modules, including digital-to-analog converters (DACs), MAC units, analog shift-and-add, and analog-to-digital converters (ADCs), reuse one set of local capacitors inside the array, performing in situ computation to save area and enhance accuracy. A compact 8.5-bit dual-threshold time-domain ADC power gates the main path most of the time, leading to a significant energy reduction. Our 65-nm prototype achieves the highest weight storage density of 559 Kb/mm2 and exceptional robustness to temperature and voltage variations (<inline-formula xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink"> <tex-math notation="LaTeX">${-} 40~^{\circ } $ </tex-math></inline-formula> C to <inline-formula xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink"> <tex-math notation="LaTeX">$105~{^{\circ }}$ </tex-math></inline-formula> C and 0.65 to 1.2 V) among SRAM-based analog CIM designs.

IEEE Journal of Solid-State CircuitsVol. 60(1)
Apple (United States) (US), Rice University (US)
Affordable and clean energy
Openalex Percentile: Top 7%
Semiconductor materials and devices
3.55
AI Navigator

Ask Laika to Summarize, Analyze, and Connect papers live on the map.

Summarize Papers & Methodologies

Extract key findings, datasets, and comparative methods across publications.

Benchmark Rankings & Visual Analytics

Rank top research institutions, authors, funders, topics, and journals by Field-Weighted Citation Impact (FWCI) and paper volume with instant charts.

Connect Distant Disciplines

Bridge topological clusters on the map to find hidden collaborative intersections.