본문

서브메뉴

Architecture and Circuit Design Optimization for Compute-In-Memory
Architecture and Circuit Design Optimization for Compute-In-Memory
Architecture and Circuit Design Optimization for Compute-In-Memory

상세정보

자료유형  
 학위논문 서양
최종처리일시  
20260202105537
ISBN  
9798263392802
DDC  
620
저자명  
Jiang, Hongwu.
서명/저자  
Architecture and Circuit Design Optimization for Compute-In-Memory
발행사항  
[Sl] : Georgia Institute of Technology, 2022
발행사항  
Ann Arbor : ProQuest Dissertations & Theses, 2022
형태사항  
126 p
주기사항  
Source: Dissertations Abstracts International, Volume: 87-05, Section: A.
주기사항  
Advisor: Yu, Shimeng.
학위논문주기  
Thesis (Ph.D.)--Georgia Institute of Technology, 2022.
초록/해제  
요약The main objective of this thesis is to optimize computing-in-memory (CIM) design for accelerating Deep Neural Network (DNN) algorithms. As compute peripheries such as analog-to-digital converters (ADCs) introduce significant overhead in CIM inference design, the first part of the research focuses on circuit optimizations for inmemory computing. In the first work, we comprehensively explore the tradeoffs involving different types of ADCs and investigate a new ADC design especially suited for the CIM, which performs the analog shift-add for multiple weight significance bits, improving the throughput and energy efficiency under similar area constraints. In the second work, we propose a resistive random access memory (RRAM) based ADC-free in-memory compute scheme validated with a prototype chip in TSMC 40nm process, which can significantly improve the hardware performance over the conventional CIM designs while achieving near-software classification accuracy on ImageNet and CIFAR-10/-100 dataset.In the second part of the thesis, the research focuses on hardware support for CIM on-chip training. To maximize hardware reuse of CIM weight stationary dataflow, we propose the CIM training architectures with the transpose weight mapping strategy. The cell design and periphery circuitry are modified to support bi-directional computing efficiently. A novel solution of signed number multiplication is also proposed to handle the negative inputs in backpropagation. Based on the silicon measurement data on a two-way SRAM-based prototype chip in TSMC 28nm process, we comprehensively explore the hardware performance for the entire SRAM-based architecture for DNN on-chip training.
일반주제명  
Silicon
일반주제명  
Design optimization
일반주제명  
Random access memory
일반주제명  
Deep learning
일반주제명  
Back propagation
일반주제명  
Macros
일반주제명  
Semiconductors
일반주제명  
Neural networks
일반주제명  
Circuits
일반주제명  
Energy efficiency
일반주제명  
CMOS
일반주제명  
Breakdowns
일반주제명  
Digitization
일반주제명  
Transistors
일반주제명  
Energy consumption
일반주제명  
Linear algebra
일반주제명  
Computer science
일반주제명  
Electrical engineering
일반주제명  
Sustainability
기타저자  
Georgia Institute of Technology.
기본자료저록  
Dissertations Abstracts International. 87-05A.
전자적 위치 및 접속  
로그인 후 원문을 볼 수 있습니다.

MARC

 008260126s2022        us                              c    eng  d
■001000017360504
■00520260202105537
■006m          o    d                
■007cr#unu||||||||
■020    ▼a9798263392802
■035    ▼a(MiAaPQ)AAI32314896
■035    ▼a(MiAaPQ)GeorgiaTech70137
■040    ▼aMiAaPQ▼cMiAaPQ
■0820  ▼a620
■1001  ▼aJiang,  Hongwu.
■24510▼aArchitecture  and  Circuit  Design  Optimization  for  Compute-In-Memory
■260    ▼a[Sl]▼bGeorgia  Institute  of  Technology▼c2022
■260  1▼aAnn  Arbor▼bProQuest  Dissertations  &  Theses▼c2022
■300    ▼a126  p
■500    ▼aSource:  Dissertations  Abstracts  International,  Volume:  87-05,  Section:  A.
■500    ▼aAdvisor:  Yu,  Shimeng.
■5021  ▼aThesis  (Ph.D.)--Georgia  Institute  of  Technology,  2022.
■520    ▼aThe  main  objective  of  this  thesis  is  to  optimize  computing-in-memory  (CIM)  design  for  accelerating  Deep  Neural  Network  (DNN)  algorithms.  As  compute  peripheries  such  as  analog-to-digital  converters  (ADCs)  introduce  significant  overhead  in  CIM  inference  design,  the  first  part  of  the  research  focuses  on  circuit  optimizations  for  inmemory  computing.  In  the  first  work,  we  comprehensively  explore  the  tradeoffs  involving  different  types  of  ADCs  and  investigate  a  new  ADC  design  especially  suited  for  the  CIM,  which  performs  the  analog  shift-add  for  multiple  weight  significance  bits,  improving  the  throughput  and  energy  efficiency  under  similar  area  constraints.  In  the  second  work,  we  propose  a  resistive  random  access  memory  (RRAM)  based  ADC-free  in-memory  compute  scheme  validated  with  a  prototype  chip  in  TSMC  40nm  process,  which  can  significantly  improve  the  hardware  performance  over  the  conventional  CIM  designs  while  achieving  near-software  classification  accuracy  on  ImageNet  and  CIFAR-10/-100  dataset.In  the  second  part  of  the  thesis,  the  research  focuses  on  hardware  support  for  CIM  on-chip  training.  To  maximize  hardware  reuse  of  CIM  weight  stationary  dataflow,  we  propose  the  CIM  training  architectures  with  the  transpose  weight  mapping  strategy.  The  cell  design  and  periphery  circuitry  are  modified  to  support  bi-directional  computing  efficiently.  A  novel  solution  of  signed  number  multiplication  is  also  proposed  to  handle  the  negative  inputs  in  backpropagation.  Based  on  the  silicon  measurement  data  on  a  two-way  SRAM-based  prototype  chip  in  TSMC  28nm  process,  we  comprehensively  explore  the  hardware  performance  for  the  entire  SRAM-based  architecture  for  DNN  on-chip  training.
■590    ▼aSchool  code:  0078.
■650  4▼aSilicon
■650  4▼aDesign  optimization
■650  4▼aRandom  access  memory
■650  4▼aDeep  learning
■650  4▼aBack  propagation
■650  4▼aMacros
■650  4▼aSemiconductors
■650  4▼aNeural  networks
■650  4▼aCircuits
■650  4▼aEnergy  efficiency
■650  4▼aCMOS
■650  4▼aBreakdowns
■650  4▼aDigitization
■650  4▼aTransistors
■650  4▼aEnergy  consumption
■650  4▼aLinear  algebra
■650  4▼aComputer  science
■650  4▼aElectrical  engineering
■650  4▼aSustainability
■690    ▼a0800
■690    ▼a0984
■690    ▼a0544
■690    ▼a0640
■71020▼aGeorgia  Institute  of  Technology.
■7730  ▼tDissertations  Abstracts  International▼g87-05A.
■790    ▼a0078
■791    ▼aPh.D.
■792    ▼a2022
■793    ▼aEnglish
■85640▼uhttp://www.riss.kr/pdu/ddodLink.do?id=T17360504▼nKERIS▼z이  자료의  원문은  한국교육학술정보원에서  제공합니다.

미리보기

내보내기

chatGPT토론

Ai 추천 관련 도서


    신착도서 더보기
    최근 3년간 통계입니다.

    소장정보

    • 예약
    • 소재불명신고
    • 나의폴더
    • 우선정리요청
    • 비도서대출신청
    • 야간 도서대출신청
    소장자료
    등록번호 청구기호 소장처 대출가능여부 대출정보
    TF14632 전자도서 대출가능 마이폴더 부재도서신고 비도서대출신청 야간 도서대출신청

    * 대출중인 자료에 한하여 예약이 가능합니다. 예약을 원하시면 예약버튼을 클릭하십시오.

    해당 도서를 다른 이용자가 함께 대출한 도서

    관련 인기도서

    로그인 후 이용 가능합니다.