A FerroFET-Based In-Memory Processor for Solving Distributed and Iterative Optimizations via Least-Squares Method

Insik Yoon; Asif Khan; Suman Datta; Arijit Raychowdhury; Muya Chang; Kai Ni; Matthew Jerry; Samantak Gangopadhyay; Gus Henry Smith; Tomer Hamam; Justin Romberg; Vijaykrishnan Narayanan

doi:10.1109/JXCDC.2019.2930222

A FerroFET-Based In-Memory Processor for Solving Distributed and Iterative Optimizations via Least-Squares Method

Insik Yoon, Asif Khan, Suman Datta, Arijit Raychowdhury, Muya Chang, Kai Ni, Matthew Jerry, Samantak Gangopadhyay, Gus Henry Smith, Tomer Hamam, Justin Romberg, Vijaykrishnan Narayanan

Research output: Contribution to journal › Article

1 Scopus citations

Abstract

In recent years, several designs that use in-memory processing to accelerate machine-learning inference problems have been proposed. Such designs are also a perfect fit for discrete, dynamic, and distributed systems that can solve large-dimensional optimization problems using iterative algorithms. For in-memory computations, ferroelectric field-effect transistors (FerroFETs) owing to their compact area and distinguishable multiple states offer promising possibilities. We present a distributed architecture that uses FerroFET memory and implements in-memory processing to solve a template problem of least squares minimization. Through this architecture, we demonstrate an improvement of 21 × in energy efficiency and 3 × in compute time compared to a static random access memory (SRAM)-based processing-in-memory (PIM) architecture.

Original language	English (US)
Article number	8767985
Pages (from-to)	132-141
Number of pages	10
Journal	IEEE Journal on Exploratory Solid-State Computational Devices and Circuits
Volume	5
Issue number	2
DOIs	https://doi.org/10.1109/JXCDC.2019.2930222
State	Published - Dec 2019

All Science Journal Classification (ASJC) codes

Electronic, Optical and Magnetic Materials
Hardware and Architecture
Electrical and Electronic Engineering

Access to Document

10.1109/JXCDC.2019.2930222

Fingerprint Dive into the research topics of 'A FerroFET-Based In-Memory Processor for Solving Distributed and Iterative Optimizations via Least-Squares Method'. Together they form a unique fingerprint.

Cite this

Yoon, I., Khan, A., Datta, S., Raychowdhury, A., Chang, M., Ni, K., Jerry, M., Gangopadhyay, S., Smith, G. H., Hamam, T., Romberg, J., & Narayanan, V. (2019). A FerroFET-Based In-Memory Processor for Solving Distributed and Iterative Optimizations via Least-Squares Method. IEEE Journal on Exploratory Solid-State Computational Devices and Circuits, 5(2), 132-141. [8767985]. https://doi.org/10.1109/JXCDC.2019.2930222

Yoon, Insik ; Khan, Asif ; Datta, Suman ; Raychowdhury, Arijit ; Chang, Muya ; Ni, Kai ; Jerry, Matthew ; Gangopadhyay, Samantak ; Smith, Gus Henry ; Hamam, Tomer ; Romberg, Justin ; Narayanan, Vijaykrishnan. / A FerroFET-Based In-Memory Processor for Solving Distributed and Iterative Optimizations via Least-Squares Method. In: IEEE Journal on Exploratory Solid-State Computational Devices and Circuits. 2019 ; Vol. 5, No. 2. pp. 132-141.

@article{8cd174a9552346a88270635828e32865,

title = "A FerroFET-Based In-Memory Processor for Solving Distributed and Iterative Optimizations via Least-Squares Method",

abstract = "In recent years, several designs that use in-memory processing to accelerate machine-learning inference problems have been proposed. Such designs are also a perfect fit for discrete, dynamic, and distributed systems that can solve large-dimensional optimization problems using iterative algorithms. For in-memory computations, ferroelectric field-effect transistors (FerroFETs) owing to their compact area and distinguishable multiple states offer promising possibilities. We present a distributed architecture that uses FerroFET memory and implements in-memory processing to solve a template problem of least squares minimization. Through this architecture, we demonstrate an improvement of 21 × in energy efficiency and 3 × in compute time compared to a static random access memory (SRAM)-based processing-in-memory (PIM) architecture.",

author = "Insik Yoon and Asif Khan and Suman Datta and Arijit Raychowdhury and Muya Chang and Kai Ni and Matthew Jerry and Samantak Gangopadhyay and Smith, {Gus Henry} and Tomer Hamam and Justin Romberg and Vijaykrishnan Narayanan",

year = "2019",

month = dec,

doi = "10.1109/JXCDC.2019.2930222",

language = "English (US)",

volume = "5",

pages = "132--141",

journal = "IEEE Journal on Exploratory Solid-State Computational Devices and Circuits",

issn = "2329-9231",

publisher = "Institute of Electrical and Electronics Engineers Inc.",

number = "2",

}

Yoon, I, Khan, A, Datta, S, Raychowdhury, A, Chang, M, Ni, K, Jerry, M, Gangopadhyay, S, Smith, GH, Hamam, T, Romberg, J & Narayanan, V 2019, 'A FerroFET-Based In-Memory Processor for Solving Distributed and Iterative Optimizations via Least-Squares Method', IEEE Journal on Exploratory Solid-State Computational Devices and Circuits, vol. 5, no. 2, 8767985, pp. 132-141. https://doi.org/10.1109/JXCDC.2019.2930222

A FerroFET-Based In-Memory Processor for Solving Distributed and Iterative Optimizations via Least-Squares Method. / Yoon, Insik; Khan, Asif; Datta, Suman; Raychowdhury, Arijit; Chang, Muya; Ni, Kai; Jerry, Matthew; Gangopadhyay, Samantak; Smith, Gus Henry; Hamam, Tomer; Romberg, Justin; Narayanan, Vijaykrishnan.

In: IEEE Journal on Exploratory Solid-State Computational Devices and Circuits, Vol. 5, No. 2, 8767985, 12.2019, p. 132-141.

Research output: Contribution to journal › Article

TY - JOUR

T1 - A FerroFET-Based In-Memory Processor for Solving Distributed and Iterative Optimizations via Least-Squares Method

AU - Yoon, Insik

AU - Khan, Asif

AU - Datta, Suman

AU - Raychowdhury, Arijit

AU - Chang, Muya

AU - Ni, Kai

AU - Jerry, Matthew

AU - Gangopadhyay, Samantak

AU - Smith, Gus Henry

AU - Hamam, Tomer

AU - Romberg, Justin

AU - Narayanan, Vijaykrishnan

PY - 2019/12

Y1 - 2019/12

N2 - In recent years, several designs that use in-memory processing to accelerate machine-learning inference problems have been proposed. Such designs are also a perfect fit for discrete, dynamic, and distributed systems that can solve large-dimensional optimization problems using iterative algorithms. For in-memory computations, ferroelectric field-effect transistors (FerroFETs) owing to their compact area and distinguishable multiple states offer promising possibilities. We present a distributed architecture that uses FerroFET memory and implements in-memory processing to solve a template problem of least squares minimization. Through this architecture, we demonstrate an improvement of 21 × in energy efficiency and 3 × in compute time compared to a static random access memory (SRAM)-based processing-in-memory (PIM) architecture.

AB - In recent years, several designs that use in-memory processing to accelerate machine-learning inference problems have been proposed. Such designs are also a perfect fit for discrete, dynamic, and distributed systems that can solve large-dimensional optimization problems using iterative algorithms. For in-memory computations, ferroelectric field-effect transistors (FerroFETs) owing to their compact area and distinguishable multiple states offer promising possibilities. We present a distributed architecture that uses FerroFET memory and implements in-memory processing to solve a template problem of least squares minimization. Through this architecture, we demonstrate an improvement of 21 × in energy efficiency and 3 × in compute time compared to a static random access memory (SRAM)-based processing-in-memory (PIM) architecture.

UR - http://www.scopus.com/inward/record.url?scp=85069901619&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=85069901619&partnerID=8YFLogxK

U2 - 10.1109/JXCDC.2019.2930222

DO - 10.1109/JXCDC.2019.2930222

M3 - Article

AN - SCOPUS:85069901619

VL - 5

SP - 132

EP - 141

JO - IEEE Journal on Exploratory Solid-State Computational Devices and Circuits

JF - IEEE Journal on Exploratory Solid-State Computational Devices and Circuits

SN - 2329-9231

IS - 2

M1 - 8767985

ER -