High-performance 3D Unstructured Mesh Deformation Using Rank Structured Matrix Computations

被引:2
|
作者
Alomairy, Rabab [1 ]
Bader, Wael [2 ]
Ltaief, Hatem [1 ]
Mesri, Youssef [2 ]
Keyes, David [1 ]
机构
[1] King Abdullah Univ Sci & Technol, Extreme Comp Res Ctr, 4700 King Abdullah Blvd, Jeddah 23955, Saudi Arabia
[2] PSL Univ, Ctr Mat Forming, MINES ParisTech, 60 Blvd St Michel, F-75272 Paris, France
关键词
3D mesh deformation; radial basis functions; hydrodynamics; low-rank matrix approximation; high-performance computing; COVID-19; FACTORIZATION; ALGORITHMS; SOLVER;
D O I
10.1145/3512756
中图分类号
TP301 [理论、方法];
学科分类号
081202 ;
摘要
The Radial Basis Function (RBF) technique is an interpolation method that produces high-quality unstructured adaptive meshes. However, the RBF-based boundary problem necessitates solving a large dense linear system with cubic arithmetic complexity that is computationally expensive and prohibitive in terms of memory footprint. In this article, we accelerate the computations of 3D unstructured mesh deformation based on RBF interpolations by exploiting the rank structured property of the matrix operator. The main idea consists in approximating the matrix off-diagonal tiles up to an application-dependent accuracy threshold. We highlight the robustness of our multiscale solver by assessing its numerical accuracy using realistic 3D geometries. In particular, we model the 3D mesh deformation on a population of the novel coronaviruses. We report and compare performance results on various parallel systems against existing state-of-the-art matrix solvers.
引用
收藏
页数:23
相关论文
共 50 条
  • [1] A general performance model of structured and unstructured mesh particle transport computations
    Mathis, MM
    Kerbyson, DJ
    [J]. JOURNAL OF SUPERCOMPUTING, 2005, 34 (02): : 181 - 199
  • [2] A General Performance Model of Structured and Unstructured Mesh Particle Transport Computations
    Mark M. Mathis
    Darren J. Kerbyson
    [J]. The Journal of Supercomputing, 2005, 34 : 181 - 199
  • [3] Anisotropic mesh adaptation for 3D flows on structured and unstructured grids
    Tam, A
    Ait-Ali-Yahia, D
    Robichaud, MP
    Moore, M
    Kozel, V
    Habashi, WG
    [J]. COMPUTER METHODS IN APPLIED MECHANICS AND ENGINEERING, 2000, 189 (04) : 1205 - 1230
  • [4] Explorable Mesh Deformation Subspaces from Unstructured 3D Generative Models
    Maesumi, Arman
    Guerrero, Paul
    Kim, Vladimir G.
    Fisher, Matthew
    Chaudhuri, Siddhartha
    Aigerman, Noam
    Ritchie, Daniel
    [J]. PROCEEDINGS OF THE SIGGRAPH ASIA 2023 CONFERENCE PAPERS, 2023,
  • [5] Turbulent flow computations on 3D unstructured grids
    Manzari, MT
    Hassan, O
    Morgan, K
    Weatherill, NP
    [J]. FINITE ELEMENTS IN ANALYSIS AND DESIGN, 1998, 30 (04) : 353 - 363
  • [6] Turbulent flow computations on 3D unstructured grids
    Univ of Wales, Swansea, United Kingdom
    [J]. Finite Elem Anal Des, 4 (353-363):
  • [7] A nitrogen-doped 3D open-structured graphite nanofiber matrix for high-performance supercapacitors
    Zhou, Yongsheng
    Zhu, Yingchun
    Xue, Dongfeng
    Xu, Bingshe
    [J]. JOURNAL OF MATERIALS CHEMISTRY A, 2018, 6 (29) : 14065 - 14068
  • [8] Design Patterns for High-Performance Matrix Computations
    Son, Hoang M.
    [J]. MODELING, SIMULATION AND OPTIMIZATION OF COMPLEX PROCESSES, 2008, : 509 - 519
  • [9] Validation of computations around high-lift configurations by structured- and unstructured-mesh
    Murayama, M
    Yamamoto, K
    [J]. JOURNAL OF AIRCRAFT, 2006, 43 (02): : 395 - 406
  • [10] Validation of computations around high-lift configurations by structured- And unstructured-mesh
    Murayama, Mitsuhiro
    Yamamoto, Kazuomi
    Kobayashi, Kunihiko
    [J]. Journal of Aircraft, 1600, 43 (02): : 395 - 406