Resilience for Massively Parallel Multigrid Solvers

DOI10.1137/15M1026122zbMath1352.65626OpenAlexW2540977105MaRDI QIDQ2830612

Björn Gmeiner, Markus Q. Huber, Ulrich Rüde, Barbara I. Wohlmuth

Publication date: 28 October 2016

Published in: SIAM Journal on Scientific Computing (Search for Journal in Brave)

Full work available at URL: https://doi.org/10.1137/15m1026122

zbMATH Keywords

Dirichlet problemelliptic equationthree-dimensional problemfault tolerant algorithmsmassively parallel and asynchronous multigrid

Mathematics Subject Classification ID

Multigrid methods; domain decomposition for boundary value problems involving PDEs (65N55) Boundary value problems for second-order elliptic equations (35J25) Parallel numerical computation (65Y05)

Related Items (6)

Multigrid at Scale? ⋮ The Resiliency of Multilevel Methods on Next-Generation Computing Platforms: Probabilistic Model and Its Analysis ⋮ \textsc{waLBerla}: a block-structured high-performance framework for multiphysics simulations ⋮ Towards local-failure local-recovery in PDE frameworks: the case of linear solvers ⋮ Stochastic subspace correction methods and fault tolerance ⋮ Randomized and fault-tolerant method of subspace corrections

Cites Work

• Unnamed Item
• Unnamed Item
• Unnamed Item
• Optimization of the multigrid-convergence rate on semi-structured meshes by local Fourier analysis
• A massively parallel solver for discrete Poisson-like problems
• Fault-tolerance techniques for high-performance computing
• A generic grid interface for parallel and adaptive scientific computing. II: Implementation and tests in DUNE
• Stabilized mixed methods for the Stokes problem
• \textit{BoomerAMG}: A parallel algebraic multigrid solver and preconditioner
• A class of iterative methods for solving saddle point problems
• Tetrahedral grid refinement
• An error-resilient redundant subspace correction method
• Analysis of iterative methods for saddle point problems: a unified approach
• Numerical Analysis of Fixed Point Algorithms in the Presence of Hardware Faults
• Numerical recovery strategies for parallel resilient Krylov linear solvers
• Multigrid Techniques
• Segmental Refinement: A Multigrid Technique for Data Locality
• Multigrid methods on parallel computers—A survey of recent developments
• Algorithm-Based Fault Tolerance for Matrix Operations
• On the use of relaxation parameters in hybrid smoothers
• Hierarchical hybrid grids: data structures and core algorithms for multigrid
• Towards Textbook Efficiency for Parallel Multigrid
• Recovery Patterns for Iterative Methods in a Parallel Unstable Environment
• A linear algebraic model of algorithm-based fault tolerance
• Algorithmic Fault Tolerance Using the Lanczos Method
• On Red-Black SOR Smoothing in Multigrid
• Performance and Scalability of Hierarchical Hybrid Multigrid Solvers for Stokes Systems
• Fault Tolerant Computation with the Sparse Grid Combination Technique
• Low-Cost Concurrent Error Detection for Floating-Point Unit (FPU) Controllers

This page was built for publication: Resilience for Massively Parallel Multigrid Solvers