Technological exploration of RRAM crossbar array for matrix-vector multiplication

Peng Gu, Boxun Li, Tianqi Tang, Shimeng Yu, Yu Cao, Yu Wang, Huazhong Yang

Research output: Chapter in Book/Report/Conference proceedingConference contribution

27 Citations (Scopus)

Abstract

The matrix-vector multiplication is the key operation for many computationally intensive algorithms. In recent years, the emerging metal oxide resistive switching random access memory (RRAM) device and RRAM crossbar array have demonstrated a promising hardware realization of the analog matrix-vector multiplication with ultra-high energy efficiency. In this paper, we analyze the impact of nonlinear voltage-current relationship of RRAM devices and the interconnect resistance as well as other crossbar array parameters on the circuit performance and present a design guide. On top of that, we propose a technological exploration flow for device parameter configuration to overcome the impact of nonideal factors and achieve a better trade-off among performance, energy and reliability for each specific application. The simulation results of a support vector machine (SVM) and MNIST pattern recognition dataset show that the RRAM crossbar array-based SVM is robust to the input signal fluctuation but sensitive to the tunneling gap deviation. A further resistance resolution test presents that a 4-bit RRAM device is able to realize a recognition accuracy of ∼ 90%, indicating the physical feasibility of RRAM crossbar array-based SVM. In addition, the proposed technological exploration flow is able to achieve 10.98% improvement of recognition accuracy on the MNIST dataset and 26.4% energy savings compared with previous work.

Original languageEnglish (US)
Title of host publication20th Asia and South Pacific Design Automation Conference, ASP-DAC 2015
PublisherInstitute of Electrical and Electronics Engineers Inc.
Pages106-111
Number of pages6
ISBN (Print)9781479977925
DOIs
StatePublished - Mar 11 2015
Event2015 20th Asia and South Pacific Design Automation Conference, ASP-DAC 2015 - Chiba, Japan
Duration: Jan 19 2015Jan 22 2015

Other

Other2015 20th Asia and South Pacific Design Automation Conference, ASP-DAC 2015
CountryJapan
CityChiba
Period1/19/151/22/15

Fingerprint

Matrix-vector multiplication
Random Access
Data storage equipment
Support vector machines
Support Vector Machine
Interconnect
Energy Saving
Energy Efficiency
Computer hardware
Pattern Recognition
Pattern recognition
High Efficiency
Energy efficiency
Oxides
High Energy
Energy conservation
Deviation
Metals
Trade-offs
Voltage

ASJC Scopus subject areas

  • Computer Science Applications
  • Electrical and Electronic Engineering
  • Control and Systems Engineering
  • Modeling and Simulation

Cite this

Gu, P., Li, B., Tang, T., Yu, S., Cao, Y., Wang, Y., & Yang, H. (2015). Technological exploration of RRAM crossbar array for matrix-vector multiplication. In 20th Asia and South Pacific Design Automation Conference, ASP-DAC 2015 (pp. 106-111). [7058989] Institute of Electrical and Electronics Engineers Inc.. https://doi.org/10.1109/ASPDAC.2015.7058989

Technological exploration of RRAM crossbar array for matrix-vector multiplication. / Gu, Peng; Li, Boxun; Tang, Tianqi; Yu, Shimeng; Cao, Yu; Wang, Yu; Yang, Huazhong.

20th Asia and South Pacific Design Automation Conference, ASP-DAC 2015. Institute of Electrical and Electronics Engineers Inc., 2015. p. 106-111 7058989.

Research output: Chapter in Book/Report/Conference proceedingConference contribution

Gu, P, Li, B, Tang, T, Yu, S, Cao, Y, Wang, Y & Yang, H 2015, Technological exploration of RRAM crossbar array for matrix-vector multiplication. in 20th Asia and South Pacific Design Automation Conference, ASP-DAC 2015., 7058989, Institute of Electrical and Electronics Engineers Inc., pp. 106-111, 2015 20th Asia and South Pacific Design Automation Conference, ASP-DAC 2015, Chiba, Japan, 1/19/15. https://doi.org/10.1109/ASPDAC.2015.7058989
Gu P, Li B, Tang T, Yu S, Cao Y, Wang Y et al. Technological exploration of RRAM crossbar array for matrix-vector multiplication. In 20th Asia and South Pacific Design Automation Conference, ASP-DAC 2015. Institute of Electrical and Electronics Engineers Inc. 2015. p. 106-111. 7058989 https://doi.org/10.1109/ASPDAC.2015.7058989
Gu, Peng ; Li, Boxun ; Tang, Tianqi ; Yu, Shimeng ; Cao, Yu ; Wang, Yu ; Yang, Huazhong. / Technological exploration of RRAM crossbar array for matrix-vector multiplication. 20th Asia and South Pacific Design Automation Conference, ASP-DAC 2015. Institute of Electrical and Electronics Engineers Inc., 2015. pp. 106-111
@inproceedings{ecf91077e8204f7cad7c9524d3fc7be8,
title = "Technological exploration of RRAM crossbar array for matrix-vector multiplication",
abstract = "The matrix-vector multiplication is the key operation for many computationally intensive algorithms. In recent years, the emerging metal oxide resistive switching random access memory (RRAM) device and RRAM crossbar array have demonstrated a promising hardware realization of the analog matrix-vector multiplication with ultra-high energy efficiency. In this paper, we analyze the impact of nonlinear voltage-current relationship of RRAM devices and the interconnect resistance as well as other crossbar array parameters on the circuit performance and present a design guide. On top of that, we propose a technological exploration flow for device parameter configuration to overcome the impact of nonideal factors and achieve a better trade-off among performance, energy and reliability for each specific application. The simulation results of a support vector machine (SVM) and MNIST pattern recognition dataset show that the RRAM crossbar array-based SVM is robust to the input signal fluctuation but sensitive to the tunneling gap deviation. A further resistance resolution test presents that a 4-bit RRAM device is able to realize a recognition accuracy of ∼ 90{\%}, indicating the physical feasibility of RRAM crossbar array-based SVM. In addition, the proposed technological exploration flow is able to achieve 10.98{\%} improvement of recognition accuracy on the MNIST dataset and 26.4{\%} energy savings compared with previous work.",
author = "Peng Gu and Boxun Li and Tianqi Tang and Shimeng Yu and Yu Cao and Yu Wang and Huazhong Yang",
year = "2015",
month = "3",
day = "11",
doi = "10.1109/ASPDAC.2015.7058989",
language = "English (US)",
isbn = "9781479977925",
pages = "106--111",
booktitle = "20th Asia and South Pacific Design Automation Conference, ASP-DAC 2015",
publisher = "Institute of Electrical and Electronics Engineers Inc.",

}

TY - GEN

T1 - Technological exploration of RRAM crossbar array for matrix-vector multiplication

AU - Gu, Peng

AU - Li, Boxun

AU - Tang, Tianqi

AU - Yu, Shimeng

AU - Cao, Yu

AU - Wang, Yu

AU - Yang, Huazhong

PY - 2015/3/11

Y1 - 2015/3/11

N2 - The matrix-vector multiplication is the key operation for many computationally intensive algorithms. In recent years, the emerging metal oxide resistive switching random access memory (RRAM) device and RRAM crossbar array have demonstrated a promising hardware realization of the analog matrix-vector multiplication with ultra-high energy efficiency. In this paper, we analyze the impact of nonlinear voltage-current relationship of RRAM devices and the interconnect resistance as well as other crossbar array parameters on the circuit performance and present a design guide. On top of that, we propose a technological exploration flow for device parameter configuration to overcome the impact of nonideal factors and achieve a better trade-off among performance, energy and reliability for each specific application. The simulation results of a support vector machine (SVM) and MNIST pattern recognition dataset show that the RRAM crossbar array-based SVM is robust to the input signal fluctuation but sensitive to the tunneling gap deviation. A further resistance resolution test presents that a 4-bit RRAM device is able to realize a recognition accuracy of ∼ 90%, indicating the physical feasibility of RRAM crossbar array-based SVM. In addition, the proposed technological exploration flow is able to achieve 10.98% improvement of recognition accuracy on the MNIST dataset and 26.4% energy savings compared with previous work.

AB - The matrix-vector multiplication is the key operation for many computationally intensive algorithms. In recent years, the emerging metal oxide resistive switching random access memory (RRAM) device and RRAM crossbar array have demonstrated a promising hardware realization of the analog matrix-vector multiplication with ultra-high energy efficiency. In this paper, we analyze the impact of nonlinear voltage-current relationship of RRAM devices and the interconnect resistance as well as other crossbar array parameters on the circuit performance and present a design guide. On top of that, we propose a technological exploration flow for device parameter configuration to overcome the impact of nonideal factors and achieve a better trade-off among performance, energy and reliability for each specific application. The simulation results of a support vector machine (SVM) and MNIST pattern recognition dataset show that the RRAM crossbar array-based SVM is robust to the input signal fluctuation but sensitive to the tunneling gap deviation. A further resistance resolution test presents that a 4-bit RRAM device is able to realize a recognition accuracy of ∼ 90%, indicating the physical feasibility of RRAM crossbar array-based SVM. In addition, the proposed technological exploration flow is able to achieve 10.98% improvement of recognition accuracy on the MNIST dataset and 26.4% energy savings compared with previous work.

UR - http://www.scopus.com/inward/record.url?scp=84926459123&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=84926459123&partnerID=8YFLogxK

U2 - 10.1109/ASPDAC.2015.7058989

DO - 10.1109/ASPDAC.2015.7058989

M3 - Conference contribution

SN - 9781479977925

SP - 106

EP - 111

BT - 20th Asia and South Pacific Design Automation Conference, ASP-DAC 2015

PB - Institute of Electrical and Electronics Engineers Inc.

ER -