본문

서브메뉴

Dispersion Analysis, Time Parallelization, and GPU Autotuning for Finite Element Methods
Dispersion Analysis, Time Parallelization, and GPU Autotuning for Finite Element Methods
Dispersion Analysis, Time Parallelization, and GPU Autotuning for Finite Element Methods

Detailed Information

자료유형  
 학위논문 서양
최종처리일시  
20260202105657
ISBN  
9798263307226
DDC  
510
저자명  
Christensen, Nicholas J.
서명/저자  
Dispersion Analysis, Time Parallelization, and GPU Autotuning for Finite Element Methods
발행사항  
[Sl] : University of Illinois at Urbana-Champaign, 2024
발행사항  
Ann Arbor : ProQuest Dissertations & Theses, 2024
형태사항  
115 p
주기사항  
Source: Dissertations Abstracts International, Volume: 87-05, Section: B.
주기사항  
Advisor: Fischer, Paul.
학위논문주기  
Thesis (Ph.D.)--University of Illinois at Urbana-Champaign, 2024.
초록/해제  
요약This dissertation explores strategies to enhance the accuracy and computational performance of finite element methods. Specifically, we analyze the dispersive error of the spectral element method, investigate a novel least-squares parallel-in-time formulation, and propose an autotuning approach for a discontinuous Galerkin finite element solver.We investigate the dispersion properties of the spectral element method (SEM) when applied to advection and advection-diffusion problems on a 1D periodic domain. Our analysis spans both the well-resolved (asymptotic) limit and the marginally resolved (pre-asymptotic) limit. To achieve this, we systematically explore a wide range of parameters, including wave numbers, element counts, and local polynomial orders. We observe that high-order methods demand fewer points-per-wavelength (PPW) than low-order methods to meet engineering tolerances during long time-integration. Remarkably, Gottlieb's observation that polynomial-based spectral methods require approximately PPW 5 for engineering tolerances holds true across various polynomial orders and element counts. Comparing use of exact quadrature on the Gauss-Legendre points and inexact quadrature (employing a diagonal mass matrix) on the Gauss-Lobotto-Legendre points, we find inexact quadrature does not significantly compromise solution accuracies at high polynomial orders.At high polynomial orders (N 4), we observe error spikes near specific values of PPW disrupt the convergence behavior, regardless of the quadrature method used. These spikes arise from previously identified gaps in the eigenvalue spectrum of the discrete operators, leading to unrepresentable phase velocities. We demonstrate that diffusive mechanisms - whether introduced numerically (via time-relaxation or an upwind discontinuous Galerkin formulation) or arising naturally from the physics of the problem - can largely mitigate these error spikes. A two-dimensional model problem further illustrates the effectiveness of the mitigation strategy.We propose MG-HLS-PinT, a novel parallel-in-time method based on multigrid principles. Derived from a normal-equations formulation of a semi-discrete partial differential equation (PDE), this approach shows potential in accelerating the solution of hyperbolic PDEs. However, it currently demands high processor counts and problems with stringent accuracy demands. We explore several methods of speeding up the approach and identify promising avenues for future exploration.We propose sub-batching as a method of speeding up the computation of large fused batched Einstein summation (einsum) GPU kernels in the context of the MIRGE-Com simulation library. Sub-batching limits the number of einsums computed concurrently and limits contention for local memory and cache. We find sub-batching enables significant performance improvements in fused batched einsum kernels compared to the baseline performance. 
일반주제명  
Mathematics
일반주제명  
Computer science
일반주제명  
Theoretical mathematics
키워드  
Finite element
키워드  
Spectral element method
키워드  
Dispersion analysis
키워드  
Autotuning
키워드  
Eigenvalue avoidance
키워드  
Einstein summation
기타저자  
University of Illinois at Urbana-Champaign Computer Science
기본자료저록  
Dissertations Abstracts International. 87-05B.
전자적 위치 및 접속  
로그인 후 원문을 볼 수 있습니다.

MARC

 008260126s2024        us                              c    eng  d
■001000017361046
■00520260202105657
■006m          o    d                
■007cr#unu||||||||
■020    ▼a9798263307226
■035    ▼a(MiAaPQ)AAI32409793
■035    ▼a(MiAaPQ)124325
■040    ▼aMiAaPQ▼cMiAaPQ
■0820  ▼a510
■1001  ▼aChristensen,  Nicholas  J.
■24510▼aDispersion  Analysis,  Time  Parallelization,  and  GPU  Autotuning  for  Finite  Element  Methods
■260    ▼a[Sl]▼bUniversity  of  Illinois  at  Urbana-Champaign▼c2024
■260  1▼aAnn  Arbor▼bProQuest  Dissertations  &  Theses▼c2024
■300    ▼a115  p
■500    ▼aSource:  Dissertations  Abstracts  International,  Volume:  87-05,  Section:  B.
■500    ▼aAdvisor:  Fischer,  Paul.
■5021  ▼aThesis  (Ph.D.)--University  of  Illinois  at  Urbana-Champaign,  2024.
■520    ▼aThis  dissertation  explores  strategies  to  enhance  the  accuracy  and  computational  performance  of  finite  element  methods.  Specifically,  we  analyze  the  dispersive  error  of  the  spectral  element  method,  investigate  a  novel  least-squares  parallel-in-time  formulation,  and  propose  an  autotuning  approach  for  a  discontinuous  Galerkin  finite  element  solver.We  investigate  the  dispersion  properties  of  the  spectral  element  method  (SEM)  when  applied  to  advection  and  advection-diffusion  problems  on  a  1D  periodic  domain.  Our  analysis  spans  both  the  well-resolved  (asymptotic)  limit  and  the  marginally  resolved  (pre-asymptotic)  limit.  To  achieve  this,  we  systematically  explore  a  wide  range  of  parameters,  including  wave  numbers,  element  counts,  and  local  polynomial  orders.  We  observe  that  high-order  methods  demand  fewer  points-per-wavelength  (PPW)  than  low-order  methods  to  meet  engineering  tolerances  during  long  time-integration.  Remarkably,  Gottlieb's  observation  that  polynomial-based  spectral  methods  require  approximately  PPW    5  for  engineering  tolerances  holds  true  across  various  polynomial  orders  and  element  counts.  Comparing  use  of  exact  quadrature  on  the  Gauss-Legendre  points  and  inexact  quadrature  (employing  a  diagonal  mass  matrix)  on  the  Gauss-Lobotto-Legendre  points,  we  find  inexact  quadrature  does  not  significantly  compromise  solution  accuracies  at  high  polynomial  orders.At  high  polynomial  orders  (N    4),  we  observe  error  spikes  near  specific  values  of  PPW  disrupt  the  convergence  behavior,  regardless  of  the  quadrature  method  used.  These  spikes  arise  from  previously  identified  gaps  in  the  eigenvalue  spectrum  of  the  discrete  operators,  leading  to  unrepresentable  phase  velocities.  We  demonstrate  that  diffusive  mechanisms  -  whether  introduced  numerically  (via  time-relaxation  or  an  upwind  discontinuous  Galerkin  formulation)  or  arising  naturally  from  the  physics  of  the  problem  -  can  largely  mitigate  these  error  spikes.  A  two-dimensional  model  problem  further  illustrates  the  effectiveness  of  the  mitigation  strategy.We  propose  MG-HLS-PinT,  a  novel  parallel-in-time  method  based  on  multigrid  principles.  Derived  from  a  normal-equations  formulation  of  a  semi-discrete  partial  differential  equation  (PDE),  this  approach  shows  potential  in  accelerating  the  solution  of  hyperbolic  PDEs.  However,  it  currently  demands  high  processor  counts  and  problems  with  stringent  accuracy  demands.  We  explore  several  methods  of  speeding  up  the  approach  and  identify  promising  avenues  for  future  exploration.We  propose  sub-batching  as  a  method  of  speeding  up  the  computation  of  large  fused  batched  Einstein  summation  (einsum)  GPU  kernels  in  the  context  of  the  MIRGE-Com  simulation  library.  Sub-batching  limits  the  number  of  einsums  computed  concurrently  and  limits  contention  for  local  memory  and  cache.  We  find  sub-batching  enables  significant  performance  improvements  in  fused  batched  einsum  kernels  compared  to  the  baseline  performance. 
■590    ▼aSchool  code:  0090.
■650  4▼aMathematics
■650  4▼aComputer  science
■650  4▼aTheoretical  mathematics
■653    ▼aFinite  element
■653    ▼aSpectral  element  method
■653    ▼aDispersion  analysis
■653    ▼aAutotuning
■653    ▼aEigenvalue  avoidance
■653    ▼aEinstein  summation
■690    ▼a0984
■690    ▼a0405
■690    ▼a0642
■71020▼aUniversity  of  Illinois  at  Urbana-Champaign▼bComputer  Science.
■7730  ▼tDissertations  Abstracts  International▼g87-05B.
■790    ▼a0090
■791    ▼aPh.D.
■792    ▼a2024
■793    ▼aEnglish
■85640▼uhttp://www.riss.kr/pdu/ddodLink.do?id=T17361046▼nKERIS▼z이  자료의  원문은  한국교육학술정보원에서  제공합니다.

Preview

Export

ChatGPT Discussion

AI Recommended Related Books


    New Books MORE
    Statistics for the past 3 years. Go to brief

    Подробнее информация.

    • Бронирование
    • не существует
    • моя папка
    • Первый запрос зрения
    • Non-Book Loan Application
    • Nighttime Book Loan Application
    материал
    Reg No. Количество платежных Местоположение статус Ленд информации
    TF16524 전자도서 대출가능 My Folder 부재도서신고 비도서대출신청 야간 도서대출신청

    * Бронирование доступны в заимствований книги. Чтобы сделать предварительный заказ, пожалуйста, нажмите кнопку бронирование

    Books borrowed together with this book

    Related Popular Books

    Available after logging in.