Dependability of Future Edge-AI Processors: Pandora's Box

Manil Dev Gomony; Anteneh Gebregiorgis; Moritz Fieback; Marc Geilen; Sander Stuijk; Jan Richter-Brockmann; Rajendra Bishnoi; Lara Arche Andradas; Sven Argo; Tim Güneysu; Mottaqiallah Taouil; Henk Corporaal; Said Hamdioui

doi:10.1109/ETS56758.2023.10174180

Dependability of Future Edge-AI Processors: Pandora's Box

Manil Dev Gomony, Anteneh Gebregiorgis, Moritz Fieback, Marc Geilen, Sander Stuijk, Jan Richter-Brockmann, Rajendra Bishnoi, Lara Arche Andradas, Sven Argo, Tim Güneysu, Mottaqiallah Taouil, Henk Corporaal, Said Hamdioui

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

7 Downloads (Pure)

Abstract

This paper addresses one of the directions of the HORIZON EU CONVOLVE project being dependability of smart edge processors based on computation-in-memory and emerging memristor devices such as RRAM. It discusses how how this alternative computing paradigm will change the way we used to do manufacturing test. In addition, it describes how these emerging devices inherently suffering from many non-idealities are calling for new solutions in order to ensure accurate and reliable edge computing. Moreover, the paper also covers the security aspects for future edge processors and shows the challenges and the future directions.

Original language	English
Title of host publication	2023 IEEE European Test Symposium (ETS)
Publisher	Institute of Electrical and Electronics Engineers
Number of pages	6
ISBN (Electronic)	979-8-3503-3634-4
DOIs	https://doi.org/10.1109/ETS56758.2023.10174180
Publication status	Published - 12 Jul 2023
Event	28th IEEE European Test Symposium, ETS 2023 - Venice, Italy Duration: 22 May 2023 → 26 May 2023

Conference

Conference	28th IEEE European Test Symposium, ETS 2023
Country/Territory	Italy
City	Venice
Period	22/05/23 → 26/05/23

Funding

This work is funded by EU’s Horizon Europe research and innovation programme under grant agreement No. 101070374.

Funders	Funder number
European Union's Horizon 2020 - Research and Innovation Framework Programme	101070374

Keywords

DSE
SoC
ULP
approximate computing
compiler stack
dynamic DL
edge-AI
memristor

Access to Document

10.1109/ETS56758.2023.10174180

pdfFinal published version, 1.38 MBLicence: TAVERNE

Cite this

Gomony, M. D., Gebregiorgis, A., Fieback, M., Geilen, M., Stuijk, S., Richter-Brockmann, J., Bishnoi, R., Andradas, L. A., Argo, S., Güneysu, T., Taouil, M., Corporaal, H., & Hamdioui, S. (2023). Dependability of Future Edge-AI Processors: Pandora's Box. In 2023 IEEE European Test Symposium (ETS) Article 10174180 Institute of Electrical and Electronics Engineers. https://doi.org/10.1109/ETS56758.2023.10174180

@inproceedings{3dabb669a50b4cf19cb1f3a38472e279,

title = "Dependability of Future Edge-AI Processors: Pandora's Box",

abstract = "This paper addresses one of the directions of the HORIZON EU CONVOLVE project being dependability of smart edge processors based on computation-in-memory and emerging memristor devices such as RRAM. It discusses how how this alternative computing paradigm will change the way we used to do manufacturing test. In addition, it describes how these emerging devices inherently suffering from many non-idealities are calling for new solutions in order to ensure accurate and reliable edge computing. Moreover, the paper also covers the security aspects for future edge processors and shows the challenges and the future directions.",

keywords = "DSE, SoC, ULP, approximate computing, compiler stack, dynamic DL, edge-AI, memristor",

author = "Gomony, {Manil Dev} and Anteneh Gebregiorgis and Moritz Fieback and Marc Geilen and Sander Stuijk and Jan Richter-Brockmann and Rajendra Bishnoi and Andradas, {Lara Arche} and Sven Argo and Tim G{\"u}neysu and Mottaqiallah Taouil and Henk Corporaal and Said Hamdioui",

year = "2023",

month = jul,

day = "12",

doi = "10.1109/ETS56758.2023.10174180",

language = "English",

booktitle = "2023 IEEE European Test Symposium (ETS)",

publisher = "Institute of Electrical and Electronics Engineers",

address = "United States",

note = "28th IEEE European Test Symposium, ETS 2023 ; Conference date: 22-05-2023 Through 26-05-2023",

}

Gomony, MD, Gebregiorgis, A, Fieback, M, Geilen, M , Stuijk, S, Richter-Brockmann, J, Bishnoi, R, Andradas, LA, Argo, S, Güneysu, T, Taouil, M, Corporaal, H & Hamdioui, S 2023, Dependability of Future Edge-AI Processors: Pandora's Box. in 2023 IEEE European Test Symposium (ETS)., 10174180, Institute of Electrical and Electronics Engineers, 28th IEEE European Test Symposium, ETS 2023, Venice, Italy, 22/05/23. https://doi.org/10.1109/ETS56758.2023.10174180

TY - GEN

T1 - Dependability of Future Edge-AI Processors: Pandora's Box

AU - Gomony, Manil Dev

AU - Gebregiorgis, Anteneh

AU - Fieback, Moritz

AU - Geilen, Marc

AU - Stuijk, Sander

AU - Richter-Brockmann, Jan

AU - Bishnoi, Rajendra

AU - Andradas, Lara Arche

AU - Argo, Sven

AU - Güneysu, Tim

AU - Taouil, Mottaqiallah

AU - Corporaal, Henk

AU - Hamdioui, Said

PY - 2023/7/12

Y1 - 2023/7/12

N2 - This paper addresses one of the directions of the HORIZON EU CONVOLVE project being dependability of smart edge processors based on computation-in-memory and emerging memristor devices such as RRAM. It discusses how how this alternative computing paradigm will change the way we used to do manufacturing test. In addition, it describes how these emerging devices inherently suffering from many non-idealities are calling for new solutions in order to ensure accurate and reliable edge computing. Moreover, the paper also covers the security aspects for future edge processors and shows the challenges and the future directions.

AB - This paper addresses one of the directions of the HORIZON EU CONVOLVE project being dependability of smart edge processors based on computation-in-memory and emerging memristor devices such as RRAM. It discusses how how this alternative computing paradigm will change the way we used to do manufacturing test. In addition, it describes how these emerging devices inherently suffering from many non-idealities are calling for new solutions in order to ensure accurate and reliable edge computing. Moreover, the paper also covers the security aspects for future edge processors and shows the challenges and the future directions.

KW - DSE

KW - SoC

KW - ULP

KW - approximate computing

KW - compiler stack

KW - dynamic DL

KW - edge-AI

KW - memristor

UR - http://www.scopus.com/inward/record.url?scp=85166208417&partnerID=8YFLogxK

U2 - 10.1109/ETS56758.2023.10174180

DO - 10.1109/ETS56758.2023.10174180

M3 - Conference contribution

BT - 2023 IEEE European Test Symposium (ETS)

PB - Institute of Electrical and Electronics Engineers

T2 - 28th IEEE European Test Symposium, ETS 2023

Y2 - 22 May 2023 through 26 May 2023

ER -

Dependability of Future Edge-AI Processors: Pandora's Box

Abstract

Conference

Funding

Keywords

Access to Document

Other files and links

Fingerprint

Cite this