Resistive RAM endurance : array-level characterization and correction techniques targeting deep learning applications

Limited endurance of resistive RAM (RRAM) is a major challenge for future computing systems. Using thorough endurance tests that incorporate fine-grained read operations at the array level, we quantify for the first time temporary write failures (TWFs) caused by intrinsic RRAM cycle-to-cycle and cel...

Full description

Saved in:

Bibliographic Details
Main Authors:	Grossi, Alessandro, Vianello, Elisa, Mohamed M. Sabry, Barlas, Marios, Grenouillet, Laurent, Coignus, Jean, Beigne, Edith, Wu, Tony, Le, Binh Q, Wootters, Mary K., Zambelli, Cristian, Nowak, Etienne, Mitra, Subhasish
Other Authors:	School of Computer Science and Engineering
Format:	Article
Language:	English
Published:	2020
Subjects:	Engineering::Computer science and engineering Characterization Deep Learning
Online Access:	https://hdl.handle.net/10356/143255
Tags:	Add Tag No Tags, Be the first to tag this record!
Institution:	Nanyang Technological University
Language:	English

id	sg-ntu-dr.10356-143255
record_format	dspace
spelling	sg-ntu-dr.10356-1432552020-08-17T04:49:00Z Resistive RAM endurance : array-level characterization and correction techniques targeting deep learning applications Grossi, Alessandro Vianello, Elisa Mohamed M. Sabry Barlas, Marios Grenouillet, Laurent Coignus, Jean Beigne, Edith Wu, Tony Le, Binh Q Wootters, Mary K. Zambelli, Cristian Nowak, Etienne Mitra, Subhasish School of Computer Science and Engineering Engineering::Computer science and engineering Characterization Deep Learning Limited endurance of resistive RAM (RRAM) is a major challenge for future computing systems. Using thorough endurance tests that incorporate fine-grained read operations at the array level, we quantify for the first time temporary write failures (TWFs) caused by intrinsic RRAM cycle-to-cycle and cell-to-cell variations. We also quantify permanent write failures (PWFs) caused by irreversible breakdown/dissolution of the conductive filament. We show how technology-, RRAM programing-, and system resilience-level solutions can be effectively combined to design new generations of energy-efficient computing systems that can successfully run deep learning (and other machine learning) applications despite TWFs and PWFs. We analyze corresponding system lifetimes and TWF bit error ratio. Nanyang Technological University Accepted version This work was supported in part by Defense Advanced Research Projects Agency, in part by NSF-SRC/Nanoelectronics Research Initiative/Global Research Collaboration Energy-Efficient Computing: From Devices to Architectures, in part by STARnet SONIC, in part by NSF, in part by the Stanford SystemX Alliance, in part by NTU Startup Grant (M4082035), and in part by the AME Programmatic Hardware Software Co-Optimization for Deep Learning (M4070301). The review of this paper was arranged by Editor Y.-H. Shih. (Corresponding author: Mohamed M. Sabry.) 2020-08-17T04:49:00Z 2020-08-17T04:49:00Z 2019 Journal Article Grossi, A., Vianello, E., Mohamed M. Sabry, Barlas, M., Grenouillet, L., Coignus, J., ... Mitra, S. (2019). Resistive RAM endurance : array-level characterization and correction techniques targeting deep learning applications. IEEE Transactions on Electron Devices, 66(3), 1281-1288. doi:10.1109/ted.2019.2894387 0018-9383 https://hdl.handle.net/10356/143255 10.1109/TED.2019.2894387 2-s2.0-85062258844 3 66 1281 1288 en M4082035 IEEE Transactions on Electron Devices © 2019 IEEE. Personal use of this material is permitted. Permission from IEEE must be obtained for all other uses, in any current or future media, including reprinting/republishing this material for advertising or promotional purposes, creating new collective works, for resale or redistribution to servers or lists, or reuse of any copyrighted component of this work in other works. The published version is available at: https://doi.org/10.1109/TED.2019.2894387. application/pdf
institution	Nanyang Technological University
building	NTU Library
country	Singapore
collection	DR-NTU
language	English
topic	Engineering::Computer science and engineering Characterization Deep Learning
spellingShingle	Engineering::Computer science and engineering Characterization Deep Learning Grossi, Alessandro Vianello, Elisa Mohamed M. Sabry Barlas, Marios Grenouillet, Laurent Coignus, Jean Beigne, Edith Wu, Tony Le, Binh Q Wootters, Mary K. Zambelli, Cristian Nowak, Etienne Mitra, Subhasish Resistive RAM endurance : array-level characterization and correction techniques targeting deep learning applications
description	Limited endurance of resistive RAM (RRAM) is a major challenge for future computing systems. Using thorough endurance tests that incorporate fine-grained read operations at the array level, we quantify for the first time temporary write failures (TWFs) caused by intrinsic RRAM cycle-to-cycle and cell-to-cell variations. We also quantify permanent write failures (PWFs) caused by irreversible breakdown/dissolution of the conductive filament. We show how technology-, RRAM programing-, and system resilience-level solutions can be effectively combined to design new generations of energy-efficient computing systems that can successfully run deep learning (and other machine learning) applications despite TWFs and PWFs. We analyze corresponding system lifetimes and TWF bit error ratio.
author2	School of Computer Science and Engineering
author_facet	School of Computer Science and Engineering Grossi, Alessandro Vianello, Elisa Mohamed M. Sabry Barlas, Marios Grenouillet, Laurent Coignus, Jean Beigne, Edith Wu, Tony Le, Binh Q Wootters, Mary K. Zambelli, Cristian Nowak, Etienne Mitra, Subhasish
format	Article
author	Grossi, Alessandro Vianello, Elisa Mohamed M. Sabry Barlas, Marios Grenouillet, Laurent Coignus, Jean Beigne, Edith Wu, Tony Le, Binh Q Wootters, Mary K. Zambelli, Cristian Nowak, Etienne Mitra, Subhasish
author_sort	Grossi, Alessandro
title	Resistive RAM endurance : array-level characterization and correction techniques targeting deep learning applications
title_short	Resistive RAM endurance : array-level characterization and correction techniques targeting deep learning applications
title_full	Resistive RAM endurance : array-level characterization and correction techniques targeting deep learning applications
title_fullStr	Resistive RAM endurance : array-level characterization and correction techniques targeting deep learning applications
title_full_unstemmed	Resistive RAM endurance : array-level characterization and correction techniques targeting deep learning applications
title_sort	resistive ram endurance : array-level characterization and correction techniques targeting deep learning applications
publishDate	2020
url	https://hdl.handle.net/10356/143255
_version_	1681057643863474176

Resistive RAM endurance : array-level characterization and correction techniques targeting deep learning applications

Similar Items