Multiscale Modeling for Application-Oriented Optimization of Resistive Random-Access Memory

Memristor-based neuromorphic systems have been proposed as a promising alternative to von Neumann computing architectures, which are currently challenged by the ever-increasing computational power required by modern artificial intelligence (AI) algorithms. The design and optimization of memristive d...

Full description

Bibliographic Details
Main Authors: Paolo La Torraca, Francesco Maria Puglisi, Andrea Padovani, Luca Larcher
Format: Article
Language:English
Published: MDPI AG 2019-10-01
Series:Materials
Subjects:
ai
Online Access:https://www.mdpi.com/1996-1944/12/21/3461
id doaj-27e00775cdd44a908f9888313ff88e07
record_format Article
spelling doaj-27e00775cdd44a908f9888313ff88e072020-11-25T02:01:24ZengMDPI AGMaterials1996-19442019-10-011221346110.3390/ma12213461ma12213461Multiscale Modeling for Application-Oriented Optimization of Resistive Random-Access MemoryPaolo La Torraca0Francesco Maria Puglisi1Andrea Padovani2Luca Larcher3Dipartimento di Scienze e Metodi dell’Ingegneria, Università di Modena e Reggio Emilia, Via Amendola 2, 42122 Reggio Emilia, ItalyDipartimento di Ingegneria “Enzo Ferrari”, Università di Modena e Reggio Emilia, Via P. Vivarelli 10/1, 41125 Modena, ItalyApplied Materials, Via Sicilia 32, 42122 Reggio Emilia, ItalyDipartimento di Scienze e Metodi dell’Ingegneria, Università di Modena e Reggio Emilia, Via Amendola 2, 42122 Reggio Emilia, ItalyMemristor-based neuromorphic systems have been proposed as a promising alternative to von Neumann computing architectures, which are currently challenged by the ever-increasing computational power required by modern artificial intelligence (AI) algorithms. The design and optimization of memristive devices for specific AI applications is thus of paramount importance, but still extremely complex, as many different physical mechanisms and their interactions have to be accounted for, which are, in many cases, not fully understood. The high complexity of the physical mechanisms involved and their partial comprehension are currently hampering the development of memristive devices and preventing their optimization. In this work, we tackle the application-oriented optimization of Resistive Random-Access Memory (RRAM) devices using a multiscale modeling platform. The considered platform includes all the involved physical mechanisms (i.e., charge transport and trapping, and ion generation, diffusion, and recombination) and accounts for the 3D electric and temperature field in the device. Thanks to its multiscale nature, the modeling platform allows RRAM devices to be simulated and the microscopic physical mechanisms involved to be investigated, the device performance to be connected to the material’s microscopic properties and geometries, the device electrical characteristics to be predicted, the effect of the forming conditions (i.e., temperature, compliance current, and voltage stress) on the device’s performance and variability to be evaluated, the analog resistance switching to be optimized, and the device’s reliability and failure causes to be investigated. The discussion of the presented simulation results provides useful insights for supporting the application-oriented optimization of RRAM technology according to specific AI applications, for the implementation of either non-volatile memories, deep neural networks, or spiking neural networks.https://www.mdpi.com/1996-1944/12/21/3461aineuromorphic computingmultiscale modelingmemristoroptimizationrramsimulation
collection DOAJ
language English
format Article
sources DOAJ
author Paolo La Torraca
Francesco Maria Puglisi
Andrea Padovani
Luca Larcher
spellingShingle Paolo La Torraca
Francesco Maria Puglisi
Andrea Padovani
Luca Larcher
Multiscale Modeling for Application-Oriented Optimization of Resistive Random-Access Memory
Materials
ai
neuromorphic computing
multiscale modeling
memristor
optimization
rram
simulation
author_facet Paolo La Torraca
Francesco Maria Puglisi
Andrea Padovani
Luca Larcher
author_sort Paolo La Torraca
title Multiscale Modeling for Application-Oriented Optimization of Resistive Random-Access Memory
title_short Multiscale Modeling for Application-Oriented Optimization of Resistive Random-Access Memory
title_full Multiscale Modeling for Application-Oriented Optimization of Resistive Random-Access Memory
title_fullStr Multiscale Modeling for Application-Oriented Optimization of Resistive Random-Access Memory
title_full_unstemmed Multiscale Modeling for Application-Oriented Optimization of Resistive Random-Access Memory
title_sort multiscale modeling for application-oriented optimization of resistive random-access memory
publisher MDPI AG
series Materials
issn 1996-1944
publishDate 2019-10-01
description Memristor-based neuromorphic systems have been proposed as a promising alternative to von Neumann computing architectures, which are currently challenged by the ever-increasing computational power required by modern artificial intelligence (AI) algorithms. The design and optimization of memristive devices for specific AI applications is thus of paramount importance, but still extremely complex, as many different physical mechanisms and their interactions have to be accounted for, which are, in many cases, not fully understood. The high complexity of the physical mechanisms involved and their partial comprehension are currently hampering the development of memristive devices and preventing their optimization. In this work, we tackle the application-oriented optimization of Resistive Random-Access Memory (RRAM) devices using a multiscale modeling platform. The considered platform includes all the involved physical mechanisms (i.e., charge transport and trapping, and ion generation, diffusion, and recombination) and accounts for the 3D electric and temperature field in the device. Thanks to its multiscale nature, the modeling platform allows RRAM devices to be simulated and the microscopic physical mechanisms involved to be investigated, the device performance to be connected to the material’s microscopic properties and geometries, the device electrical characteristics to be predicted, the effect of the forming conditions (i.e., temperature, compliance current, and voltage stress) on the device’s performance and variability to be evaluated, the analog resistance switching to be optimized, and the device’s reliability and failure causes to be investigated. The discussion of the presented simulation results provides useful insights for supporting the application-oriented optimization of RRAM technology according to specific AI applications, for the implementation of either non-volatile memories, deep neural networks, or spiking neural networks.
topic ai
neuromorphic computing
multiscale modeling
memristor
optimization
rram
simulation
url https://www.mdpi.com/1996-1944/12/21/3461
work_keys_str_mv AT paololatorraca multiscalemodelingforapplicationorientedoptimizationofresistiverandomaccessmemory
AT francescomariapuglisi multiscalemodelingforapplicationorientedoptimizationofresistiverandomaccessmemory
AT andreapadovani multiscalemodelingforapplicationorientedoptimizationofresistiverandomaccessmemory
AT lucalarcher multiscalemodelingforapplicationorientedoptimizationofresistiverandomaccessmemory
_version_ 1724957196241862656