TY - JOUR
T1 - A Study of the Effect of RRAM Reliability Soft Errors on the Performance of RRAM-Based Neuromorphic Systems
AU - Tosson, Amr M.S.
AU - Yu, Shimeng
AU - Anis, Mohab H.
AU - Wei, Lan
N1 - Funding Information:
Mohab H. Anis (S’98–M’03) received the B.Sc. degree (Hons.) in electronics and communication engineering from Cairo University, Cairo, Egypt, in 1997, and the M.A.Sc. and Ph.D. degrees in electrical engineering from the University of Waterloo, Waterloo, ON, Canada, in 1999 and 2003, respectively. He is currently an Associate Professor of elec-tronics and communications engineering with The American University in Cairo, Cairo, and is also an Adjunct Professor with the University of Waterloo. He has authored/co-authored over 80 articles in the ACM/IEEE journals and conferences and is the Author of the books Multi-Threshold CMOS Digital Circuits-Managing Leakage Power (Kluwer, 2003) and Low-Power Design of Nanometer FPGAs: Architecture and EDA (Morgan Kaufmann, 2009). Dr. Anis received the 2009 Early Research Award from the Ministry of Research and Innovation, Canada, the 2004 Douglas R. Colton Medal for Research Excellence, (in recognition of excellence in research leading to new understanding and novel developments in Microsystems, Canada), and the 2002 International Low-Power Design Prize. He is also the TPC-Chair of the IEEE International Conference on Microelectronics from 2007 to 2009, and has been a member of the Program Committee for several IEEE conferences. He is on the Editorial Board of the IEEE TRANSACTIONS ON CIRCUITS AND SYSTEMS-II, the Microelectronics Journal, the Journal of Circuits, Systems and Computers, and the ASP Journal of Low Power Electronics.
Publisher Copyright:
© 2017 IEEE.
PY - 2017/11
Y1 - 2017/11
N2 - Resistive RAM (RRAM) device has been extensively used as a scalable nonvolatile memory cell in neuromorphic systems due to its several advantages, including its small size and low-power requirements. However, resulting from the stochastic nature of the oxygen vacancies, the RRAM device suffers from reliability soft errors. In this paper, we provide for the first time a modeling framework to compute the effect of those soft errors on the system accuracy. Applying the proposed technique on a case-study system used to recognize the MNIST data set, our simulation results show that the system accuracy can degrade from 91.6% to 43% due to the RRAM reliability soft errors. To overcome this loss in the system performance, various possible adjustments to the parameters of the neuron pulses are analyzed. Furthermore, in this paper, two methodologies are proposed for automatically detecting and fixing the degradation in the system accuracy caused by the RRAM reliability soft errors. Using the suggested methodologies, the system accuracy of our case-study system can be restored back from 43% to 91.6% with small increase in the training cycle duration and with as small as 0.1% increment in the energy consumption of the system.
AB - Resistive RAM (RRAM) device has been extensively used as a scalable nonvolatile memory cell in neuromorphic systems due to its several advantages, including its small size and low-power requirements. However, resulting from the stochastic nature of the oxygen vacancies, the RRAM device suffers from reliability soft errors. In this paper, we provide for the first time a modeling framework to compute the effect of those soft errors on the system accuracy. Applying the proposed technique on a case-study system used to recognize the MNIST data set, our simulation results show that the system accuracy can degrade from 91.6% to 43% due to the RRAM reliability soft errors. To overcome this loss in the system performance, various possible adjustments to the parameters of the neuron pulses are analyzed. Furthermore, in this paper, two methodologies are proposed for automatically detecting and fixing the degradation in the system accuracy caused by the RRAM reliability soft errors. Using the suggested methodologies, the system accuracy of our case-study system can be restored back from 43% to 91.6% with small increase in the training cycle duration and with as small as 0.1% increment in the energy consumption of the system.
KW - MNIST recognition system
KW - RRAM reliability soft errors
KW - neuromorphic systems
KW - nonvolatile memory
KW - resistive RAM (RRAM) arrays
UR - http://www.scopus.com/inward/record.url?scp=85028450040&partnerID=8YFLogxK
UR - http://www.scopus.com/inward/citedby.url?scp=85028450040&partnerID=8YFLogxK
U2 - 10.1109/TVLSI.2017.2734819
DO - 10.1109/TVLSI.2017.2734819
M3 - Article
AN - SCOPUS:85028450040
SN - 1063-8210
VL - 25
SP - 3125
EP - 3137
JO - IEEE Transactions on Very Large Scale Integration (VLSI) Systems
JF - IEEE Transactions on Very Large Scale Integration (VLSI) Systems
IS - 11
M1 - 8010884
ER -