본문

서브메뉴

Hardware-Software Codesign for Accelerating Graph Neural Networks on FPGA
Hardware-Software Codesign for Accelerating Graph Neural Networks on FPGA
Hardware-Software Codesign for Accelerating Graph Neural Networks on FPGA

상세정보

자료유형  
 학위논문 서양
최종처리일시  
20250211152728
ISBN  
9798383689035
DDC  
621.3
저자명  
Zhang, Bingyi.
서명/저자  
Hardware-Software Codesign for Accelerating Graph Neural Networks on FPGA
발행사항  
[Sl] : University of Southern California, 2024
발행사항  
Ann Arbor : ProQuest Dissertations & Theses, 2024
형태사항  
156 p
주기사항  
Source: Dissertations Abstracts International, Volume: 86-02, Section: B.
주기사항  
Advisor: Prasanna, Viktor.
학위논문주기  
Thesis (Ph.D.)--University of Southern California, 2024.
초록/해제  
요약Graph Neural Networks (GNNs) have revolutionized many real-world applications where data can be represented as graphs. These applications include recommendation systems, social networks, traffic prediction, computer vision tasks (e.g., the tasks in autonomous driving), etc. Many applications require high-performance execution (e.g., low latency or high throughput) of GNNs. To this end, several state-of-the-art libraries (e.g., PyTorch Geometric and Deep Graph Library) have been developed for executing GNNs on general-purpose processors, including CPUs and GPGPUs. However, current GNN libraries on general-purpose processors achieve sub-optimal performance due to several challenges: 1. Irregular data structures: graphs in real-world applications are highly unstructured, with uneven degree distribution. Such irregularity leads to complex data access patterns. 2. Heterogeneous computation kernels: GNNs involve both sparse computation kernels (e.g., sparse-dense matrix multiplication) and dense computation kernels (dense-dense matrix multiplication). While general-purpose processors are efficient for dense computations, their data path and memory hierarchy are inefficient for sparse computations. 3. Dynamic data sparsity: In many applications, the graph connectivity and the data sparsity of vertex features are unknown before executing the GNN model. Therefore, a GNN system needs to deal with the data sparsity of the graphs dynamically. Such dynamic data sparsity makes it difficult for the compiler and runtime system to generate an optimal execution scheme for GNNs. 4. Mixture of models: Some GNN-based applications use the combined strength of different machine learning models. For example, GNN-based computer vision tasks utilize a mixture of convolutional neural networks (CNNs) and GNN models. Such a combination leads to complex data flow.In this dissertation, we address the above challenges through novel hardware-software codesigns on Field Programmable Gate Array (FPGA). First, to address the challenges of irregular data structures and heterogeneous computation kernels, we develop a hardware-software codesign on FPGA for GNN inference, named GraphAGILE. GraphAGILE incorporates a compiler and an accelerator design on FPGA. For a given input GNN model and an input graph, the compiler translates them into the intermediate representation (IR). Then, the compiler performs several compiler optimizations and generates a sequence of instructions for hardware execution. The hardware accelerator on FPGA executes various computation kernels through flexible customized data paths and memory organization. Second, we propose Dynasparse, an efficient codesign of runtime system and hardware to exploit the dynamic sparsity in GNN inference. The hardware design of Dynasparse has a flexible data path to execute the computation kernels of various data sparsity. The runtime system utilizes a theoretical performance model to dynamically map a GNN computation kernel to the computation primitive based on data sparsity. Third, we propose GCV-Turbo, a hardware-software codesign accelerating GNN-based computer vision (CV) tasks. GNN-based CV tasks involve a mixture of GNN layers and CNN layers. To this end, GCV-Turbo incorporates novel compiler optimizations to orchestrate the dataflow of two types of models. GCV-Turbo further incorporates novel hardware mechanisms to facilitate efficient data layout transformation between different layers. Our implementations based on our codesign methodology achieve superior performance on various GNN-based applications.
일반주제명  
Computer engineering
일반주제명  
Computer science
키워드  
Compiler
키워드  
Computer architecture
키워드  
Graph Neural Network
키워드  
Runtime
키워드  
Autonomous driving
기타저자  
University of Southern California Computer Engineering
기본자료저록  
Dissertations Abstracts International. 86-02B.
전자적 위치 및 접속  
로그인 후 원문을 볼 수 있습니다.

MARC

 008250123s2024        us                              c    eng  d
■001000017163591
■00520250211152728
■006m          o    d                
■007cr#unu||||||||
■020    ▼a9798383689035
■035    ▼a(MiAaPQ)AAI31490494
■040    ▼aMiAaPQ▼cMiAaPQ
■0820  ▼a621.3
■1001  ▼aZhang,  Bingyi.
■24510▼aHardware-Software  Codesign  for  Accelerating  Graph  Neural  Networks  on  FPGA
■260    ▼a[Sl]▼bUniversity  of  Southern  California▼c2024
■260  1▼aAnn  Arbor▼bProQuest  Dissertations  &  Theses▼c2024
■300    ▼a156  p
■500    ▼aSource:  Dissertations  Abstracts  International,  Volume:  86-02,  Section:  B.
■500    ▼aAdvisor:  Prasanna,  Viktor.
■5021  ▼aThesis  (Ph.D.)--University  of  Southern  California,  2024.
■520    ▼aGraph  Neural  Networks  (GNNs)  have  revolutionized  many  real-world  applications  where  data  can  be  represented  as  graphs.  These  applications  include  recommendation  systems,  social  networks,  traffic  prediction,  computer  vision  tasks  (e.g.,  the  tasks  in  autonomous  driving),  etc.  Many  applications  require  high-performance  execution  (e.g.,  low  latency  or  high  throughput)  of  GNNs.  To  this  end,  several  state-of-the-art  libraries  (e.g.,  PyTorch  Geometric  and  Deep  Graph  Library)  have  been  developed  for  executing  GNNs  on  general-purpose  processors,  including  CPUs  and  GPGPUs.  However,  current  GNN  libraries  on  general-purpose  processors  achieve  sub-optimal  performance  due  to  several  challenges:  1.  Irregular  data  structures:  graphs  in  real-world  applications  are  highly  unstructured,  with  uneven  degree  distribution.  Such  irregularity  leads  to  complex  data  access  patterns.  2.  Heterogeneous  computation  kernels:  GNNs  involve  both  sparse  computation  kernels  (e.g.,  sparse-dense  matrix  multiplication)  and  dense  computation  kernels  (dense-dense  matrix  multiplication).  While  general-purpose  processors  are  efficient  for  dense  computations,  their  data  path  and  memory  hierarchy  are  inefficient  for  sparse  computations.  3.  Dynamic  data  sparsity:  In  many  applications,  the  graph  connectivity  and  the  data  sparsity  of  vertex  features  are  unknown  before  executing  the  GNN  model.  Therefore,  a  GNN  system  needs  to  deal  with  the  data  sparsity  of  the  graphs  dynamically.  Such  dynamic  data  sparsity  makes  it  difficult  for  the  compiler  and  runtime  system  to  generate  an  optimal  execution  scheme  for  GNNs.  4.  Mixture  of  models:  Some  GNN-based  applications  use  the  combined  strength  of  different  machine  learning  models.  For  example,  GNN-based  computer  vision  tasks  utilize  a  mixture  of  convolutional  neural  networks  (CNNs)  and  GNN  models.  Such  a  combination  leads  to  complex  data  flow.In  this  dissertation,  we  address  the  above  challenges  through  novel  hardware-software  codesigns  on  Field  Programmable  Gate  Array  (FPGA).  First,  to  address  the  challenges  of  irregular  data  structures  and  heterogeneous  computation  kernels,  we  develop  a  hardware-software  codesign  on  FPGA  for  GNN  inference,  named  GraphAGILE.  GraphAGILE  incorporates  a  compiler  and  an  accelerator  design  on  FPGA.  For  a  given  input  GNN  model  and  an  input  graph,  the  compiler  translates  them  into  the  intermediate  representation  (IR).  Then,  the  compiler  performs  several  compiler  optimizations  and  generates  a  sequence  of  instructions  for  hardware  execution.  The  hardware  accelerator  on  FPGA  executes  various  computation  kernels  through  flexible  customized  data  paths  and  memory  organization.  Second,  we  propose  Dynasparse,  an  efficient  codesign  of  runtime  system  and  hardware  to  exploit  the  dynamic  sparsity  in  GNN  inference.  The  hardware  design  of  Dynasparse  has  a  flexible  data  path  to  execute  the  computation  kernels  of  various  data  sparsity.  The  runtime  system  utilizes  a  theoretical  performance  model  to  dynamically  map  a  GNN  computation  kernel  to  the  computation  primitive  based  on  data  sparsity.  Third,  we  propose  GCV-Turbo,  a  hardware-software  codesign  accelerating  GNN-based  computer  vision  (CV)  tasks.  GNN-based  CV  tasks  involve  a  mixture  of  GNN  layers  and  CNN  layers.  To  this  end,  GCV-Turbo  incorporates  novel  compiler  optimizations  to  orchestrate  the  dataflow  of  two  types  of  models.  GCV-Turbo  further  incorporates  novel  hardware  mechanisms  to  facilitate  efficient  data  layout  transformation  between  different  layers.  Our  implementations  based  on  our  codesign  methodology  achieve  superior  performance  on  various  GNN-based  applications.
■590    ▼aSchool  code:  0208.
■650  4▼aComputer  engineering
■650  4▼aComputer  science
■653    ▼aCompiler
■653    ▼aComputer  architecture
■653    ▼aGraph  Neural  Network
■653    ▼aRuntime
■653    ▼aAutonomous  driving
■690    ▼a0464
■690    ▼a0984
■71020▼aUniversity  of  Southern  California▼bComputer  Engineering.
■7730  ▼tDissertations  Abstracts  International▼g86-02B.
■790    ▼a0208
■791    ▼aPh.D.
■792    ▼a2024
■793    ▼aEnglish
■85640▼uhttp://www.riss.kr/pdu/ddodLink.do?id=T17163591▼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
    TF12920 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.