Abstract:More than a half century of Computational Fluid Dynamics / Plus d'un demi-siècle de mécanique des fluides numérique Recent advances in domain decomposition methods for large-scale saddle point problems Progrès récents dans les méthodes de décomposition de domaine pour le problème du point de selle à grande
“…• FETI method [16,40] • P.L. Lions algorithm [20] • Inexact Coarse solve [30] • Saddle point problem [31] • Boundary Element Methods [29] • Multiscale Finite Element methods [27] • Time dependent Maxwell system [5] • Least Square problems [9] • Purely algebraic settings [9,17] Note that for Helmholtz or frequency Maxwell type problems, efficient coarse spaces are more easily built when they are based on a coarse grid discretisation of the underlying variational form as it is classically done in multigrid methods, see [18] and references therein for a detailed mathematical analysis and [4] for extensive numerical tests of various approaches.…”
Section: Coarse Space Constructionsmentioning
confidence: 99%
“…In [31], we solve three dimensional elasticity problems for steel-rubber (ν = 0, 4999) structures (see Figure 5.1) discretized by a finite element method with continuous pressure with up to a billion degrees of freedom on 16,800 cores.…”
Section: Libraries For Large Scale Computationsmentioning
Scalability of parallel solvers for problems with high heterogeneities relies on adaptive coarse spaces built from generalized eigenvalue problems in the subdomains. The corresponding theory is powerful and flexible but the development of an efficient parallel implementation is challenging. We report here on recent advances in adaptive coarse spaces and on their open source implementations.
“…• FETI method [16,40] • P.L. Lions algorithm [20] • Inexact Coarse solve [30] • Saddle point problem [31] • Boundary Element Methods [29] • Multiscale Finite Element methods [27] • Time dependent Maxwell system [5] • Least Square problems [9] • Purely algebraic settings [9,17] Note that for Helmholtz or frequency Maxwell type problems, efficient coarse spaces are more easily built when they are based on a coarse grid discretisation of the underlying variational form as it is classically done in multigrid methods, see [18] and references therein for a detailed mathematical analysis and [4] for extensive numerical tests of various approaches.…”
Section: Coarse Space Constructionsmentioning
confidence: 99%
“…In [31], we solve three dimensional elasticity problems for steel-rubber (ν = 0, 4999) structures (see Figure 5.1) discretized by a finite element method with continuous pressure with up to a billion degrees of freedom on 16,800 cores.…”
Section: Libraries For Large Scale Computationsmentioning
Scalability of parallel solvers for problems with high heterogeneities relies on adaptive coarse spaces built from generalized eigenvalue problems in the subdomains. The corresponding theory is powerful and flexible but the development of an efficient parallel implementation is challenging. We report here on recent advances in adaptive coarse spaces and on their open source implementations.
In recent years, solvers for finite-element discretizations of linear or linearized saddle-point problems, like the Stokes and Oseen equations, have become well established. There are two main classes of preconditioners for such systems: those based on a block-factorization approach and those based on monolithic multigrid. Both classes of preconditioners have several critical choices to be made in their composition, such as the selection of a suitable relaxation scheme for monolithic multigrid. From existing studies, some insight can be gained as to what options are preferable in low-performance computing settings, but there are very few fair comparisons of these approaches in the literature, particularly for modern architectures, such as GPUs. In this paper, we perform a comparison between a Block-Triangular preconditioner and monolithic multigrid methods with the three most common choices of relaxation scheme – Braess-Sarazin, Vanka, and Schur-Uzawa. We develop a performant Vanka relaxation algorithm for structured-grid discretizations, which takes advantage of memory efficiencies in this setting. We detail the behavior of the various CUDA kernels for the multigrid relaxation schemes and evaluate their individual arithmetic intensity, performance, and runtime. Running a preconditioned FGMRES solver for the Stokes equations with these preconditioners allows us to compare their efficiency in a practical setting. We show that monolithic multigrid can outperform Block-Triangular preconditioning, and that using Vanka or Braess-Sarazin relaxation is most efficient. Even though multigrid with Vanka relaxation exhibits reduced performance on the CPU (up to 100% slower than Braess-Sarazin), it is able to outperform Braess-Sarazin by more than 20% on the GPU, making it a competitive algorithm, especially given the high amount of algorithmic tuning needed for effective Braess-Sarazin relaxation.
scite is a Brooklyn-based organization that helps researchers better discover and understand research articles through Smart Citations–citations that display the context of the citation and describe whether the article provides supporting or contrasting evidence. scite is used by students and researchers from around the world and is funded in part by the National Science Foundation and the National Institute on Drug Abuse of the National Institutes of Health.