Using Spin-Hall MTJs to Build an Energy-Efficient In-memory Computation Platform

Masoud Zabihi; Zhengyang Zhao; D. C. Mahendra; Zamshed I. Chowdhury; Salonik Resch; Thomas Peterson; Ulya R. Karpuzcu; Jian Ping Wang; Sachin S. Sapatnekar

doi:10.1109/ISQED.2019.8697377

Using Spin-Hall MTJs to Build an Energy-Efficient In-memory Computation Platform

Masoud Zabihi, Zhengyang Zhao, D. C. Mahendra, Zamshed I. Chowdhury, Salonik Resch, Thomas Peterson, Ulya R. Karpuzcu, Jian Ping Wang, Sachin S. Sapatnekar

Electrical and Computer Engineering

Research output: Chapter in Book/Report/Conference proceeding › Conference contribution

25 Scopus citations

Abstract

We present the Spin Hall Effect (SHE) Computational Random Access Memory (CRAM) for in-memory computation, incorporating considerations at the device, gate, and functional levels. For two specific applications (2-D convolution and neuromorphic digit recognition), we show that SHE-CRAM is 3x faster and has over 4x lower energy than a prior STT-based CRAM implementation, and is over 2000x faster and at least 130x more energy-efficient than state-of-the-art near-memory processing.

Original language	English (US)
Title of host publication	Proceedings of the 20th International Symposium on Quality Electronic Design, ISQED 2019
Publisher	IEEE Computer Society
Pages	52-57
Number of pages	6
ISBN (Electronic)	9781728103921
DOIs	https://doi.org/10.1109/ISQED.2019.8697377
State	Published - Apr 23 2019
Event	20th International Symposium on Quality Electronic Design, ISQED 2019 - Santa Clara, United States Duration: Mar 6 2019 → Mar 7 2019

Publication series

Name	Proceedings - International Symposium on Quality Electronic Design, ISQED
Volume	2019-March
ISSN (Print)	1948-3287
ISSN (Electronic)	1948-3295

Conference

Conference	20th International Symposium on Quality Electronic Design, ISQED 2019
Country/Territory	United States
City	Santa Clara
Period	3/6/19 → 3/7/19

Bibliographical note

Funding Information:
Accepted 31 October 2003. The authors thank the Norwegian Research Council (NFR 125693/720 and 140730/720), the Norwegian Ministry of Environment Transport and Effect Programme, and the Toxic Substances Research Initiative in Canada for funding this study. The authors thank Tine Borgen for performing the serological analysis. Current address for Andrew E. Derocher is University of Alberta, Department of Biological Sciences, T6G 2E9 Edmonton, Alberta, Canada. Address correspondence to Janneche Utne Skaare, National Veterinary Institute, The Norwegian School of Veterinary Science, PO Box 8156, Dep. N-0033 Oslo, Norway. E-mail: janneche.skaare@vetinst.no

Publisher Copyright:
© 2019 IEEE.

Keywords

In-memory computing
Neuromorphic computing
Nonvolatile memory
SHE-CRAM
Spintronics

Publisher link

10.1109/ISQED.2019.8697377

Find It

Find It at U of M Libraries

Cite this

Zabihi, M., Zhao, Z., Mahendra, D. C., Chowdhury, Z. I., Resch, S., Peterson, T., Karpuzcu, U. R., Wang, J. P., & Sapatnekar, S. S. (2019). Using Spin-Hall MTJs to Build an Energy-Efficient In-memory Computation Platform. In Proceedings of the 20th International Symposium on Quality Electronic Design, ISQED 2019 (pp. 52-57). [8697377] (Proceedings - International Symposium on Quality Electronic Design, ISQED; Vol. 2019-March). IEEE Computer Society. https://doi.org/10.1109/ISQED.2019.8697377

Using Spin-Hall MTJs to Build an Energy-Efficient In-memory Computation Platform. / Zabihi, Masoud; Zhao, Zhengyang; Mahendra, D. C. et al.

Proceedings of the 20th International Symposium on Quality Electronic Design, ISQED 2019. IEEE Computer Society, 2019. p. 52-57 8697377 (Proceedings - International Symposium on Quality Electronic Design, ISQED; Vol. 2019-March).

Research output: Chapter in Book/Report/Conference proceeding › Conference contribution

Zabihi, M, Zhao, Z, Mahendra, DC, Chowdhury, ZI, Resch, S, Peterson, T, Karpuzcu, UR , Wang, JP & Sapatnekar, SS 2019, Using Spin-Hall MTJs to Build an Energy-Efficient In-memory Computation Platform. in Proceedings of the 20th International Symposium on Quality Electronic Design, ISQED 2019., 8697377, Proceedings - International Symposium on Quality Electronic Design, ISQED, vol. 2019-March, IEEE Computer Society, pp. 52-57, 20th International Symposium on Quality Electronic Design, ISQED 2019, Santa Clara, United States, 3/6/19. https://doi.org/10.1109/ISQED.2019.8697377

Zabihi M, Zhao Z, Mahendra DC, Chowdhury ZI, Resch S, Peterson T et al. Using Spin-Hall MTJs to Build an Energy-Efficient In-memory Computation Platform. In Proceedings of the 20th International Symposium on Quality Electronic Design, ISQED 2019. IEEE Computer Society. 2019. p. 52-57. 8697377. (Proceedings - International Symposium on Quality Electronic Design, ISQED). doi: 10.1109/ISQED.2019.8697377

@inproceedings{bc8276c47cc944dc945fde5b961ca7f4,

title = "Using Spin-Hall MTJs to Build an Energy-Efficient In-memory Computation Platform",

abstract = "We present the Spin Hall Effect (SHE) Computational Random Access Memory (CRAM) for in-memory computation, incorporating considerations at the device, gate, and functional levels. For two specific applications (2-D convolution and neuromorphic digit recognition), we show that SHE-CRAM is 3x faster and has over 4x lower energy than a prior STT-based CRAM implementation, and is over 2000x faster and at least 130x more energy-efficient than state-of-the-art near-memory processing.",

keywords = "In-memory computing, Neuromorphic computing, Nonvolatile memory, SHE-CRAM, Spintronics",

author = "Masoud Zabihi and Zhengyang Zhao and Mahendra, {D. C.} and Chowdhury, {Zamshed I.} and Salonik Resch and Thomas Peterson and Karpuzcu, {Ulya R.} and Wang, {Jian Ping} and Sapatnekar, {Sachin S.}",

note = "Funding Information: Accepted 31 October 2003. The authors thank the Norwegian Research Council (NFR 125693/720 and 140730/720), the Norwegian Ministry of Environment Transport and Effect Programme, and the Toxic Substances Research Initiative in Canada for funding this study. The authors thank Tine Borgen for performing the serological analysis. Current address for Andrew E. Derocher is University of Alberta, Department of Biological Sciences, T6G 2E9 Edmonton, Alberta, Canada. Address correspondence to Janneche Utne Skaare, National Veterinary Institute, The Norwegian School of Veterinary Science, PO Box 8156, Dep. N-0033 Oslo, Norway. E-mail: janneche.skaare@vetinst.no Publisher Copyright: {\textcopyright} 2019 IEEE.; 20th International Symposium on Quality Electronic Design, ISQED 2019 ; Conference date: 06-03-2019 Through 07-03-2019",

year = "2019",

month = apr,

day = "23",

doi = "10.1109/ISQED.2019.8697377",

language = "English (US)",

series = "Proceedings - International Symposium on Quality Electronic Design, ISQED",

publisher = "IEEE Computer Society",

pages = "52--57",

booktitle = "Proceedings of the 20th International Symposium on Quality Electronic Design, ISQED 2019",

}

TY - GEN

T1 - Using Spin-Hall MTJs to Build an Energy-Efficient In-memory Computation Platform

AU - Zabihi, Masoud

AU - Zhao, Zhengyang

AU - Mahendra, D. C.

AU - Chowdhury, Zamshed I.

AU - Resch, Salonik

AU - Peterson, Thomas

AU - Karpuzcu, Ulya R.

AU - Wang, Jian Ping

AU - Sapatnekar, Sachin S.

N1 - Funding Information: Accepted 31 October 2003. The authors thank the Norwegian Research Council (NFR 125693/720 and 140730/720), the Norwegian Ministry of Environment Transport and Effect Programme, and the Toxic Substances Research Initiative in Canada for funding this study. The authors thank Tine Borgen for performing the serological analysis. Current address for Andrew E. Derocher is University of Alberta, Department of Biological Sciences, T6G 2E9 Edmonton, Alberta, Canada. Address correspondence to Janneche Utne Skaare, National Veterinary Institute, The Norwegian School of Veterinary Science, PO Box 8156, Dep. N-0033 Oslo, Norway. E-mail: janneche.skaare@vetinst.no Publisher Copyright: © 2019 IEEE.

PY - 2019/4/23

Y1 - 2019/4/23

N2 - We present the Spin Hall Effect (SHE) Computational Random Access Memory (CRAM) for in-memory computation, incorporating considerations at the device, gate, and functional levels. For two specific applications (2-D convolution and neuromorphic digit recognition), we show that SHE-CRAM is 3x faster and has over 4x lower energy than a prior STT-based CRAM implementation, and is over 2000x faster and at least 130x more energy-efficient than state-of-the-art near-memory processing.

AB - We present the Spin Hall Effect (SHE) Computational Random Access Memory (CRAM) for in-memory computation, incorporating considerations at the device, gate, and functional levels. For two specific applications (2-D convolution and neuromorphic digit recognition), we show that SHE-CRAM is 3x faster and has over 4x lower energy than a prior STT-based CRAM implementation, and is over 2000x faster and at least 130x more energy-efficient than state-of-the-art near-memory processing.

KW - In-memory computing

KW - Neuromorphic computing

KW - Nonvolatile memory

KW - SHE-CRAM

KW - Spintronics

UR - http://www.scopus.com/inward/record.url?scp=85065191788&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=85065191788&partnerID=8YFLogxK

U2 - 10.1109/ISQED.2019.8697377

DO - 10.1109/ISQED.2019.8697377

M3 - Conference contribution

AN - SCOPUS:85065191788

T3 - Proceedings - International Symposium on Quality Electronic Design, ISQED

SP - 52

EP - 57

BT - Proceedings of the 20th International Symposium on Quality Electronic Design, ISQED 2019

PB - IEEE Computer Society

T2 - 20th International Symposium on Quality Electronic Design, ISQED 2019

Y2 - 6 March 2019 through 7 March 2019

ER -

Using Spin-Hall MTJs to Build an Energy-Efficient In-memory Computation Platform

Abstract

Publication series

Conference

Bibliographical note

Keywords

Publisher link

Find It

Other files and links

Fingerprint

Cite this