Yield Evaluation of Faulty Memristive Crossbar Array-based Neural Networks with Repairability

Anu Bala; Saurabh Khandelwal; Abusaleh Jabir; Marco Ottavi

doi:10.1109/IOLTS56730.2022.9897183

Yield Evaluation of Faulty Memristive Crossbar Array-based Neural Networks with Repairability

Anu Bala^*, Saurabh Khandelwal, Abusaleh Jabir, Marco Ottavi

^*Corresponding author for this work

Research output: Chapter in Book/Report/Conference proceeding › Conference contribution › Academic › peer-review

100 Downloads (Pure)

Abstract

This paper evaluates the yield of a memristor-based crossbar array of artificial neural networks in the presence of stuck-at-faults (SAFs). A technique based on Markov chains is used to estimate the yield in the presence of stuck-at-faults. This method provides a high degree of accuracy. Another method that is used for analysis and comparison is the Poisson distribution, which uses the sum of all repairable fault patterns. A fault repair mechanism is also considered when evaluating the yield of the memristor crossbar array. The results demonstrate that the yield could be improved with redundancies and a higher repairable stuck-at-fault ratio.

Original language	English
Title of host publication	2022 IEEE 28th International Symposium on On-Line Testing and Robust System Design (IOLTS)
Editors	Alessandro Savino, Paolo Rech, Stefano Di Carlo, Dimitris Gizopoulos
Publisher	IEEE
Number of pages	5
ISBN (Electronic)	978-1-6654-7355-2
ISBN (Print)	978-1-6654-7356-9
DOIs	https://doi.org/10.1109/IOLTS56730.2022.9897183
Publication status	Published - 27 Sept 2022
Event	28th IEEE International Symposium on On-Line Testing and Robust System Design, IOLTS 2022 - Torino, Italy Duration: 12 Sept 2022 → 14 Sept 2022 Conference number: 28

Conference

Conference	28th IEEE International Symposium on On-Line Testing and Robust System Design, IOLTS 2022
Abbreviated title	IOLTS 2022
Country/Territory	Italy
City	Torino
Period	12/09/22 → 14/09/22

Keywords

Memristive Crossbar
Memristor
Neural Network
Stuck-at-Faults
22/4 OA procedure

Access to Document

10.1109/IOLTS56730.2022.9897183

Yield_Evaluation_of_Faulty_Memristive_Crossbar_Array-based_Neural_Networks_with_RepairabilityFinal published version, 1.74 MBLicence: Taverne

Cite this

@inproceedings{f59dbb398a344bb2a6f0a7e61049b996,

title = "Yield Evaluation of Faulty Memristive Crossbar Array-based Neural Networks with Repairability",

abstract = "This paper evaluates the yield of a memristor-based crossbar array of artificial neural networks in the presence of stuck-at-faults (SAFs). A technique based on Markov chains is used to estimate the yield in the presence of stuck-at-faults. This method provides a high degree of accuracy. Another method that is used for analysis and comparison is the Poisson distribution, which uses the sum of all repairable fault patterns. A fault repair mechanism is also considered when evaluating the yield of the memristor crossbar array. The results demonstrate that the yield could be improved with redundancies and a higher repairable stuck-at-fault ratio.",

keywords = "Memristive Crossbar, Memristor, Neural Network, Stuck-at-Faults, 22/4 OA procedure",

author = "Anu Bala and Saurabh Khandelwal and Abusaleh Jabir and Marco Ottavi",

note = "Publisher Copyright: {\textcopyright} 2022 IEEE.; 28th IEEE International Symposium on On-Line Testing and Robust System Design, IOLTS 2022, IOLTS 2022 ; Conference date: 12-09-2022 Through 14-09-2022",

year = "2022",

month = sep,

day = "27",

doi = "10.1109/IOLTS56730.2022.9897183",

language = "English",

isbn = "978-1-6654-7356-9",

editor = "Alessandro Savino and Paolo Rech and {Di Carlo}, Stefano and Dimitris Gizopoulos",

booktitle = "2022 IEEE 28th International Symposium on On-Line Testing and Robust System Design (IOLTS)",

publisher = "IEEE",

address = "United States",

}

Bala, A, Khandelwal, S, Jabir, A & Ottavi, M 2022, Yield Evaluation of Faulty Memristive Crossbar Array-based Neural Networks with Repairability. in A Savino, P Rech, S Di Carlo & D Gizopoulos (eds), 2022 IEEE 28th International Symposium on On-Line Testing and Robust System Design (IOLTS). IEEE, 28th IEEE International Symposium on On-Line Testing and Robust System Design, IOLTS 2022, Torino, Italy, 12/09/22. https://doi.org/10.1109/IOLTS56730.2022.9897183

Yield Evaluation of Faulty Memristive Crossbar Array-based Neural Networks with Repairability. / Bala, Anu; Khandelwal, Saurabh; Jabir, Abusaleh et al.
2022 IEEE 28th International Symposium on On-Line Testing and Robust System Design (IOLTS). ed. / Alessandro Savino; Paolo Rech; Stefano Di Carlo; Dimitris Gizopoulos. IEEE, 2022.

Research output: Chapter in Book/Report/Conference proceeding › Conference contribution › Academic › peer-review

TY - GEN

T1 - Yield Evaluation of Faulty Memristive Crossbar Array-based Neural Networks with Repairability

AU - Bala, Anu

AU - Khandelwal, Saurabh

AU - Jabir, Abusaleh

AU - Ottavi, Marco

N1 - Conference code: 28

PY - 2022/9/27

Y1 - 2022/9/27

N2 - This paper evaluates the yield of a memristor-based crossbar array of artificial neural networks in the presence of stuck-at-faults (SAFs). A technique based on Markov chains is used to estimate the yield in the presence of stuck-at-faults. This method provides a high degree of accuracy. Another method that is used for analysis and comparison is the Poisson distribution, which uses the sum of all repairable fault patterns. A fault repair mechanism is also considered when evaluating the yield of the memristor crossbar array. The results demonstrate that the yield could be improved with redundancies and a higher repairable stuck-at-fault ratio.

AB - This paper evaluates the yield of a memristor-based crossbar array of artificial neural networks in the presence of stuck-at-faults (SAFs). A technique based on Markov chains is used to estimate the yield in the presence of stuck-at-faults. This method provides a high degree of accuracy. Another method that is used for analysis and comparison is the Poisson distribution, which uses the sum of all repairable fault patterns. A fault repair mechanism is also considered when evaluating the yield of the memristor crossbar array. The results demonstrate that the yield could be improved with redundancies and a higher repairable stuck-at-fault ratio.

KW - Memristive Crossbar

KW - Memristor

KW - Neural Network

KW - Stuck-at-Faults

KW - 22/4 OA procedure

UR - http://www.scopus.com/inward/record.url?scp=85141374284&partnerID=8YFLogxK

U2 - 10.1109/IOLTS56730.2022.9897183

DO - 10.1109/IOLTS56730.2022.9897183

M3 - Conference contribution

AN - SCOPUS:85141374284

SN - 978-1-6654-7356-9

BT - 2022 IEEE 28th International Symposium on On-Line Testing and Robust System Design (IOLTS)

A2 - Savino, Alessandro

A2 - Rech, Paolo

A2 - Di Carlo, Stefano

A2 - Gizopoulos, Dimitris

PB - IEEE

T2 - 28th IEEE International Symposium on On-Line Testing and Robust System Design, IOLTS 2022

Y2 - 12 September 2022 through 14 September 2022

ER -

Yield Evaluation of Faulty Memristive Crossbar Array-based Neural Networks with Repairability

Abstract

Conference

Keywords

Access to Document

Other files and links

Fingerprint

Cite this