检索结果-南通市图书馆

XB-SIM*:A Simulation Framework for Modeling and Exploration of ReRAM-Based CNN Acceleration Design

在线全文

同方期刊数据库

学校读者我要写书评

暂无评论

tsinghua science and technology 2021年第3期26卷 322-334页

作者： Xiang Fei youhui zhang weimin zheng department of computer science and technology Tsinghua UniversityBeijing 100084China beijing National Research Center for Information science and technology Beijing 100084China

Resistive Random Access Memory(ReRAM)-based neural network accelerators have potential to surpass their digital counterparts in computational efficiency and ***,design of these accelerators faces a number of challenges including imperfections of the Re RAM device and a large amount of calculations required to accurately simulate the *** present XB-SIM,a simulation framework for Re RAM-crossbar-based Convolutional Neural Network(CNN)***-SIM can be flexibly configured to simulate the accelerator’s structure and clock-driven behaviors at the architecture *** framework also includes an Re RAM-aware Neural Network(NN)training algorithm and a CNN-oriented mapper to train an NN and map it onto the simulated design *** of the simulator has been verified by the corresponding circuit simulation of a real ***,a batch processing mode of the massive calculations that are required to mimic the behavior of Re RAM-crossbar circuits is proposed to fully apply the computational concurrency of the mapping *** CPU/GPGPU,this batch processing mode can improve the simulation speed by up to 5.02 or *** this framework,comprehensive architectural exploration and end-to-end evaluation have been achieved,which provide some insights for systemic optimization.

关键词： deep neural network Resistive Random Access Memory(Re RAM) simulation accelerator processing in memory

Towards“General Purpose”Brain-Inspired Computing System

在线全文

学校读者我要写书评

暂无评论

tsinghua science and technology 2021年第5期26卷 664-673页

作者： youhui zhang Peng Qu weimin zheng the department of computer science and technology Tsinghua UniversityBeijing 100084China

Brain-inspired computing refers to computational models,methods,and systems,that are mainly inspired by the processing mode or structure of brain.A recent study proposed the concept of"neuromorphic completeness"and the corresponding system hierarchy,which is helpful to determine the capability boundary of brain-inspired computing system and to judge whether hardware and software of brain-inspired computing are compatible with each *** a position paper,this article analyzes the existing brain-inspired chips design characteristics and the current so-called"general purpose"application development frameworks for brain-inspired computing,as well as introduces the background and the potential of this ***,some key features of this concept are presented through the comparison with the Turing completeness and approximate computation,and the analyses of the relationship with"general-purpose"brain-inspired computing systems(it means that computing systems can support all computable applications).In the end,a promising technical approach to realize such computing systems is introduced,as well as the on-going research and the work *** believe that this work is conducive to the design of extensible neuromorphic complete hardware-primitives and the corresponding *** this basis,it is expected to gradually realize"general purpose"brain-inspired computing system,in order to take into account the functionality completeness and application efficiency.

关键词： brain-inspired computing neuromorphic computing computational completeness hardware/software decoupling system hierarchy

Hardware Implementation of Spiking Neural Networks on FPGA

在线全文

学校读者我要写书评

暂无评论

tsinghua science and technology 2020年第4期25卷 479-486页

作者： Jianhui Han Zhaolin Li weimin zheng youhui zhang the Institute of Microelectronics Tsinghua UniversityBeijing 100084China the Research Institute of Information technology Tsinghua UniversityBeijing 100084China the department of computer science and technology Tsinghua UniversityBeijing 100084China

Inspired by real biological neural models,Spiking Neural Networks(SNNs)process information with discrete spikes and show great potential for building low-power neural network *** paper proposes a hardware implementation of SNN based on Field-Programmable Gate Arrays(FPGA).It features a hybrid updating algorithm,which combines the advantages of existing algorithms to simplify hardware design and improve *** proposed design supports up to 16384 neurons and 16.8 million synapses but requires minimal hardware resources and archieves a very low power consumption of 0.477 W.A test platform is built based on the proposed design using a Xilinx FPGA evaluation board,upon which we deploy a classification task on the MNIST *** evaluation results show an accuracy of 97.06%and a frame rate of 161 frames per second.

关键词： Spiking Neural Network(SNN) Field-Programmable Gate Arrays(FPGA) digital circuit low-power MNIST

Solaris系统多线程检查点设置与卷回恢复

在线全文

学校读者我要写书评

暂无评论

计算机工程与应用 2000年第8期36卷 45-47页

作者：张悠慧汪东升郑纬民清华大学计算机系北京100084

文章利用UNIX进程检查点设置思想，结合多线程在Solaris系统中的实现特点，提出了一种适合于Solaris操作系统的多线程检查点设置与恢复技术，其检查点设置与恢复技术具有在用户级实现、对用户透明和简单高效的特点。文章主要介绍检查点... 详细信息

文章利用UNIX进程检查点设置思想，结合多线程在Solaris系统中的实现特点，提出了一种适合于Solaris操作系统的多线程检查点设置与恢复技术，其检查点设置与恢复技术具有在用户级实现、对用户透明和简单高效的特点。文章主要介绍检查点信息的保存与恢复、函数换名、包裹，线程号映射等关键技术。

关键词： Solaris系统 UNIX 多线和检查点设置卷回恢复

在线全文

学校读者我要写书评

暂无评论

基于忆阻器的近似计算方法

清华大学学报（自然科学版） 2021年第6期61卷 610-617页

作者：季宇张悠慧郑纬民清华大学计算机科学与技术系北京信息科学与技术国家研究中心北京100084

忆阻器是一种非易失性存储器件,目前主要有两种方法用忆阻器实现通用计算:通过忆阻器交叉开关阵列支持神经网络来逼近任意函数;用忆阻器构造基础的门电路,再进一步实现任意Boole逻辑计算。前者存在误差难以控制的问题,后者相比传统数字... 详细信息

忆阻器是一种非易失性存储器件,目前主要有两种方法用忆阻器实现通用计算:通过忆阻器交叉开关阵列支持神经网络来逼近任意函数;用忆阻器构造基础的门电路,再进一步实现任意Boole逻辑计算。前者存在误差难以控制的问题,后者相比传统数字电路优势不显著。该文设计了一种针对忆阻器的通用近似计算范式,基于忆阻器的硬件架构通用现场可编程突触阵列(GP-FPSA),结合了两种方法的优点来实现基于忆阻器的高效且误差可控的通用近似计算。在具体设计上,充分考虑了神经网络近似能力的限制,通过万能近似器来解决直接训练神经网络误差过大且不可控的问题,并结合控制流实现了复杂函数的拆分,降低近似构造的开销,最后通过基于忆阻器的架构设计,使得通用计算能力大幅提升。

关键词：近似计算神经网络万能近似器忆阻器

在线全文

学校读者我要写书评

暂无评论

实现软硬件解耦合的类脑计算硬件设计方法

计算机研究与发展 2021年第6期58卷 1146-1154页

作者：渠鹏陈嘉杰张悠慧郑纬民清华大学计算机科学与技术系北京信息科学与技术国家研究中心北京100084 数学工程与先进计算国家重点实验室江苏无锡214125

类脑计算是一个涉及到多领域、多学科的新兴领域,对计算神经科学、人工智能和新型体系结构设计都具有重要的支持和启发意义.但是类脑计算系统领域发展所面临的重要问题之一是软硬件紧耦合.近期的一项研究提出了神经形态完备性的概念,为... 详细信息

类脑计算是一个涉及到多领域、多学科的新兴领域,对计算神经科学、人工智能和新型体系结构设计都具有重要的支持和启发意义.但是类脑计算系统领域发展所面临的重要问题之一是软硬件紧耦合.近期的一项研究提出了神经形态完备性的概念,为实现类脑计算系统领域的软硬件解耦合提供了理论支持,并作为样例研究提出了对应的系统层次结构设计.作为这一工作的后续,首先对神经形态完备性和类脑计算层次结构中部分关键的概念进行了阐述与讨论,之后进一步提出了在这一概念和体系结构设计下,实现支持软硬件解耦合的类脑计算硬件设计方法的构想,即由执行原语集合设计以及硬件实现方法设计组成的迭代调整的设计流程.最后,展示了正在进行的基于FPGA的相应评估平台工作.这一硬件设计方法有助于实现神经形态完备的高效原语集合和芯片设计,从而有利于实现类脑计算系统领域的软硬件解耦合.

关键词：类脑计算完备性软硬件解耦合 FPGA 性能评估

Design and application of new storage systems

在线全文

学校读者我要写书评

暂无评论

Frontiers of Information technology & Electronic Engineering 2023年第5期24卷 633-636页

作者： Guangyan zhang Dan FENG Keqin LI Zili SHAO Nong XIAO Jin XIONG weimin zheng department of computer science and technology Tsinghua UniversityBeijing 100084China School of computer science and technology Huazhong University of Science and TechnologyWuhan 430074China department of computer science State University of New YorkNY 12561USA department of computer science and Engineering The Chinese University of Hong KongHong KongChina School of computer science and Engineering Sun Yat-sen UniversityGuangzhou 510000China Institute of Computing technology Chinese Academy of SciencesBeijing 100190China

A storage system is the core of a computer,and plays an important role in the sustainable development of emerging strategic industries,such as artificial intelligence,big data,cloud computing,and the Internet of *** stack access is a major factor restricting the performance of data-intensive systems because of the increasing performance of processors and network devices.

关键词： computer artificial application

CEPC Technical Design Report

在线全文

学校读者我要写书评

暂无评论

Radiation Detection technology and Methods 2024年第1期8卷 I0003-I0016,1-1091页

作者： Waleed Abdallah Tiago Carlos Adorno de Freitas Konstantin Afanaciev Shakeel Ahmad Ijaz Ahmed Xiaocong Ai Abid Aleem Wolfgang Altmannshofer Fabio Alves weiming An Rui An Daniele Paolo Anderle Stefan Antusch Yasuo Arai Andrej Arbuzov Abdesslam Arhrib Mustafa Ashry Sha Bai Yu Bai Yang Bai Vipul Bairathi Csaba Balazs Philip Bambade Yong Ban Tripamo Bandyopadhyay Shou-Shan Bao Desmond P.Barber Ayse Bat Varvara Batozskaya subash Chandra Behera Alexander Belyaev Michele Bertucci Xiao-Jun Bi Yuanjie Bi Tianjian Bian Fabrizio Bianchi Thomas Biekotter Michela Biglietti Shalva Bilanishvili Deng Binglin Denis Bodrov Anton Bogomyagkov Serge Bondarenko Stewart Boogert Maarten Boonekamp Marcello Borri Angelo Bosotti Vincent Boudry Mohammed Boukidi Igor Boyko Ivanka Bozovic Giuseppe Bozzi Jean-Claude Brient Anastasiia Budzinskaya Masroor Bukhari Vladimir Bytev Giacomo Cacciapaglia Hua Cai Wenyong Cai Wujun Cai Yijian Cai Yizhou Cai Yuchen Cai Haiying Cai Huacheng Cai Lorenzo Calibbi Junsong Cang Guofu Cao Jianshe Cao Antoine Chance Xuejun Chang Yue Chang Zhe Chang Xinyuan Chang Wei Chao Auttakit Chatrabhuti Yimin Che Yuzhi Che Bin Chen Danping Chen Fuqing Chen Fusan Chen Gang Chen Guoming Chen Hua-Xing Chen Huirun Chen Jinhui Chen Ji-Yuan Chen Kai Chen Mali Chen Mingjun Chen Mingshui Chen Ning Chen Shanhong Chen Shanzhen Chen Shao-Long Chen Shaomin Chen Shiqiang Chen Tianlu Chen Wei Chen Xiang Chen Xiaoyu Chen Xin Chen Xun Chen Xurong Chen Ye Chen Ying Chen Yukai Chen Zelin Chen Zilin Chen Gang Chen Boping Chen Chunhui Chen Hok Chuen Cheng Huajie Cheng Shan Cheng Tongguang Cheng Yunlong Chi Pietro Chimenti Wen Han Chiu Guk Cho Ming-Chung Chu Xiaotong Chu Ziliang Chu Guglielmo Coloretti Andreas Crivellin Hanhua Cui Xiaohao Cui Zhaoyuan Cui Brunella D'Anzi Ling-Yun Dai Xinchen Dai Xuwen Dai Antonio De Maria Nicola De Filippis Christophe De La Taille Francesca De Mori Chiara De Sio Elisa Del Core Shuangxue Deng Wei-Tian Deng Zhi Deng Ziyan Deng Bhupal Dev Tang Dewen Biagio Di Micco Ran Ding Siqin Dingl Yadong Ding Haiyi Dong Jianin AGH university of science and technology Faculty of Physics and Applied Computer ScienceKrakow Anhui university HefeiAnhui Applied Physics department Federal Urdu University of ArtsScience and TechnologyIslamabad Banaras Hindu university Institute of ScienceVaranasi Bandirma Onyedi Eylul university Bandirma Beihang university Beijing beijing Computational science Research Center Beijing beijing Conveyi LTD Beijing beijing Glass Research Institute Co. LtdBeijing beijing Institute of technology Beijing beijing Normal university Beijing Brown university ProvidenceRI Budker Institute of Nuclear Physics Novosibirsk Cadi Ayyad university Laboratory of Fundamental and Applied PhysicsPolydisciplinary FacultyMarrakech Cairo university Giza Center for Theory and Computation National Tsing Hua UniversityHsinchu Central china Normal university WuhanHubei Central South university ChangshaHunan Centro de Fisica Teorica de Particulas Universidade Tecnica de LisboaLisboa Chengdu KaitengSifang Digital Radio&TV Equipment Co. LtdChengduSichuan china academy of engineering physics MianyangSichuan china building materials academy Beijing china Center of Advanced science and technology Beijing china Institute of Atomic Energy Beijing china Jiliang university HangzhouZhejiang china Nuclear Instrument CO. LTDBeijing china university of Geosciences Beijing china university of Geosciences WuhanHubei china university of Mining and technology XuzhouJiangsu china West Normal university NanchongSichuan Chongqing university Department of PhysicsChongqing Chulalongkorn university Bangkok City university of Hong Kong Department of PhysicsHong Kong Cockcroft Institute Daresbury Cornell university IthacaNY CPPM(AMU/IPhU&CNRS/INP) Marseille Dalian Minzu university DalianLiaoning Dalian university of technology Department of PhysicsDalian Deutsches Elektronen-Synchrotron DESY Hamburg Dzhelepov Laboratory of Nuclear Problems of Joint Institute for Nuclear Research(DLNP JINR) Dubna East china university Of science And Technol

The Circular Electron Positron Collider(CEPC)is a large scientific project initiated and hosted by china,fostered through extensive collaboration with international *** complex comprises four accelerators:a 30 GeV Linac,a 1.1 GeV Damping Ring,a Booster capable of achieving energies up to 180 GeV,and a Collider operating at varying energy modes(Z,W,H,and tt).The Linac and Damping Ring are situated on the surface,while the subterranean Booster and Collider are housed in a 100 km circumference underground tunnel,strategically accommodating future expansion with provisions for a potential super Proton Proton Collider(SPPC).The CEPC primarily serves as a Higgs *** its baseline design with synchrotron radiation(SR)power of 30 MW per beam,it can achieve a luminosity of 5×10^(34)cm^(-2)s^(-1)per interaction point(IP),resulting in an integrated luminosity of 13 ab^(-1)for two IPs over a decade,producing 2.6 million Higgs *** the SR power to 50 MW per beam expands the CEPC's capability to generate 4.3 million Higgs bosons,facilitating precise measurements of Higgs coupling at sub-percent levels,exceeding the precision expected from the HL-LHC by an order of *** Technical Design Report(TDR)follows the Preliminary Conceptual Design Report(Pre-CDR,2015)and the Conceptual Design Report(CDR,2018),comprehensively detailing the machine's layout,performance metrics,physical design and analysis,technical systems design,R&D and prototyping efforts,and associated civil engineering ***,it includes a cost estimate and a preliminary construction timeline,establishing a framework for forthcoming engineering design phase and site selection *** is anticipated to begin around 2027-2028,pending government approval,with an estimated duration of 8 *** commencement of experiments and data collection could potentially be initiated in the mid-2030s.

关键词： initiated exceeding precise

维普期刊数据库评论

在线全文

维普期刊数据库

学校读者我要写书评

暂无评论

类Paxos共识算法研究进展

计算机研究与发展 2019年第4期56卷 692-707页

作者：王江章明星武永卫陈康郑纬民清华大学计算机科学与技术系北京100084 北京信息科学与技术国家研究中心北京100084 清华大学深圳研究生院广东深圳518055

随着互联网数据量和业务量的快速增长,集群规模越来越大,由于机器和网络等故障使得业务中断的可能性越来越高.如何实现一个容错的分布式系统十分重要.多机达成共识是分布式容错系统中一个最基础,最核心的问题.Paxos等一系列共识算法的... 详细信息

随着互联网数据量和业务量的快速增长,集群规模越来越大,由于机器和网络等故障使得业务中断的可能性越来越高.如何实现一个容错的分布式系统十分重要.多机达成共识是分布式容错系统中一个最基础,最核心的问题.Paxos等一系列共识算法的出现有效地解决了这个问题.近年来,越来越多的系统使用共识相关技术,关于分布式共识算法的研究也层出不穷.这些共识算法可以被划分成2个大类:强领导者共识算法和弱领导者共识算法.随着远程直接内存访问(remote direct memory access, RDMA)等网络技术和现场可编程门阵列(field-programmable gate array, FPGA)等硬件技术的发展,又出现了一些结合新型网络和硬件技术的共识算法研究,用来提升分布式系统的性能.将从分布式共识算法发展历程的角度,介绍Paxos系列算法,阐述算法演进过程中的关键研究,讨论相关算法在不同场景下的优劣势,并展望该类算法的未来发展方向与前景.

关键词：分布式共识容错 Paxos算法强领导者共识弱领导者共识