Pub Date : 2020-06-10DOI: 10.1109/EDCC51268.2020.00013
Shihao Song, Anup Das, Nagarajan Kandasamy
As process technology continues to scale aggressively, circuit aging in a neuromorphic hardware due to negative bias temperature instability (NBTI) and time-dependent dielectric breakdown (TDDB) is becoming a critical reliability issue and is expected to proliferate when using non-volatile memory (NVM) for synaptic storage. This is because NVM devices require high voltages and currents to access their synaptic weights, which further accelerate the circuit aging in neuromorphic hardware. Current methods for qualifying reliability are overly conservative, since they estimate circuit aging considering worst-case operating conditions and unnecessarily constrain performance. This paper proposes RENEU, a reliability-oriented approach to map machine learning applications to neuromorphic hardware, with the aim of improving system-wide reliability, without compromising key performance metrics such as execution time of these applications on the hardware. Fundamental to RENEU is a novel formulation of the aging of CMOS-based circuits in a neuromorphic hardware considering different failure mechanisms. Using this formulation, RENEU develops a system- wide reliability model which can be used inside a design-space exploration framework involving the mapping of neurons and synapses to the hardware. To this end, RENEU uses an instance of Particle Swarm Optimization (PSO) to generate mappings that are Pareto-optimal in terms of performance and reliability. We evaluate RENEU using different machine learning applications on a state-of-the-art neuromorphic hardware with NVM synapses. Our results demonstrate an average 38% reduction in circuit aging, leading to an average 18% improvement in the lifetime of the hardware compared to current practices. RENEU only introduces a marginal performance overhead of 5% compared to a performance-oriented state-of-the-art.
{"title":"Improving Dependability of Neuromorphic Computing With Non-Volatile Memory","authors":"Shihao Song, Anup Das, Nagarajan Kandasamy","doi":"10.1109/EDCC51268.2020.00013","DOIUrl":"https://doi.org/10.1109/EDCC51268.2020.00013","url":null,"abstract":"As process technology continues to scale aggressively, circuit aging in a neuromorphic hardware due to negative bias temperature instability (NBTI) and time-dependent dielectric breakdown (TDDB) is becoming a critical reliability issue and is expected to proliferate when using non-volatile memory (NVM) for synaptic storage. This is because NVM devices require high voltages and currents to access their synaptic weights, which further accelerate the circuit aging in neuromorphic hardware. Current methods for qualifying reliability are overly conservative, since they estimate circuit aging considering worst-case operating conditions and unnecessarily constrain performance. This paper proposes RENEU, a reliability-oriented approach to map machine learning applications to neuromorphic hardware, with the aim of improving system-wide reliability, without compromising key performance metrics such as execution time of these applications on the hardware. Fundamental to RENEU is a novel formulation of the aging of CMOS-based circuits in a neuromorphic hardware considering different failure mechanisms. Using this formulation, RENEU develops a system- wide reliability model which can be used inside a design-space exploration framework involving the mapping of neurons and synapses to the hardware. To this end, RENEU uses an instance of Particle Swarm Optimization (PSO) to generate mappings that are Pareto-optimal in terms of performance and reliability. We evaluate RENEU using different machine learning applications on a state-of-the-art neuromorphic hardware with NVM synapses. Our results demonstrate an average 38% reduction in circuit aging, leading to an average 18% improvement in the lifetime of the hardware compared to current practices. RENEU only introduces a marginal performance overhead of 5% compared to a performance-oriented state-of-the-art.","PeriodicalId":212573,"journal":{"name":"2020 16th European Dependable Computing Conference (EDCC)","volume":"1 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2020-06-10","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"129580234","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Pub Date : 2018-07-01DOI: 10.1109/usnc-ursi.2018.8602510
Naveen Sharma, Soonwook Hwang
Members Tarek Abdelzaher, University of Illinois at Urbana-Champaign, USA Betty Cheng, Michigan State University, USA Ada Diaconescu, Telecom ParisTech, France Yixin Diao, IBM Research Simon Dobson, University of St Andrews, Scotland Holger Giese, Hasso Plattner Institute, Germany Soonwook Hwang, Korea Inst. of Science and Technology Information, South Korea Michael Kozuch, Intel, USA Philippe Lalanda, University of Grenoble, France Daniel Menasce, George Mason University, USA Arif Merchant, Google, USA Dejan Milojicic, HP Labs, USA Manish Parashar, Rutgers University, USA Hartmut Schmeck, Karlsruhe Institute of Technology, Germany Alan Sill, Texas Tech University, USA Vladimir Vlassov, KTH Royal Institute of Technology, Sweden Xiaoyun Zhu, Huawei, USA
{"title":"Steering Committee","authors":"Naveen Sharma, Soonwook Hwang","doi":"10.1109/usnc-ursi.2018.8602510","DOIUrl":"https://doi.org/10.1109/usnc-ursi.2018.8602510","url":null,"abstract":"Members Tarek Abdelzaher, University of Illinois at Urbana-Champaign, USA Betty Cheng, Michigan State University, USA Ada Diaconescu, Telecom ParisTech, France Yixin Diao, IBM Research Simon Dobson, University of St Andrews, Scotland Holger Giese, Hasso Plattner Institute, Germany Soonwook Hwang, Korea Inst. of Science and Technology Information, South Korea Michael Kozuch, Intel, USA Philippe Lalanda, University of Grenoble, France Daniel Menasce, George Mason University, USA Arif Merchant, Google, USA Dejan Milojicic, HP Labs, USA Manish Parashar, Rutgers University, USA Hartmut Schmeck, Karlsruhe Institute of Technology, Germany Alan Sill, Texas Tech University, USA Vladimir Vlassov, KTH Royal Institute of Technology, Sweden Xiaoyun Zhu, Huawei, USA","PeriodicalId":212573,"journal":{"name":"2020 16th European Dependable Computing Conference (EDCC)","volume":"31 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2018-07-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"115399043","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
京公网安备 11010802042870号
京ICP备2023020795号-1
扫码关注我们
求助内容：
应助结果提醒方式：