An error-resilient redundant subspace correction method

DOI10.1007/S00791-016-0270-6zbMATH Open1364.65275arXiv1309.0212OpenAlexW1750964885MaRDI QIDQ2359624FDOQ2359624

Authors: Tao Cui, Jinchao Xu, Chensong Zhang 

Publication date: 22 June 2017

Published in: Computing and Visualization in Science (Search for Journal in Brave)

Abstract: As we stride toward the exascale era, due to increasing complexity of supercomputers, hard and soft errors are causing more and more problems in high-performance scientific and engineering computation. In order to improve reliability (increase the mean time to failure) of computing systems, a lot of efforts have been devoted to developing techniques to forecast, prevent, and recover from errors at different levels, including architecture, application, and algorithm. In this paper, we focus on algorithmic error resilient iterative linear solvers and introduce a redundant subspace correction method. Using a general framework of redundant subspace corrections, we construct iterative methods, which have the following properties: (1) Maintain convergence when error occurs assuming it is detectable; (2) Introduce low computational overhead when no error occurs; (3) Require only small amount of local (point-to-point) communication compared to traditional methods and maintain good load balance; (4) Improve the mean time to failure. With the proposed method, we can improve reliability of many scientific and engineering applications. Preliminary numerical experiments demonstrate the efficiency and effectiveness of the new subspace correction method.

Full work available at URL: https://arxiv.org/abs/1309.0212

Recommendations

• Randomized and fault-tolerant method of subspace corrections
• Stochastic subspace correction methods and fault tolerance
• Algorithm-based error-detection schemes for iterative solution of partial differential equations
• Resilience for massively parallel multigrid solvers
• Numerical recovery strategies for parallel resilient Krylov linear solvers.

zbMATH Keywords

fault-toleranceSchwarz methodssubspace correctionerror resilience

Mathematics Subject Classification ID

Multigrid methods; domain decomposition for boundary value problems involving PDEs (65N55) Finite element, Rayleigh-Ritz and Galerkin methods for boundary value problems involving PDEs (65N30)

Cites Work

• A Fast and High Quality Multilevel Scheme for Partitioning Irregular Graphs
• Finite Element Methods for Maxwell's Equations
• Title not available (Why is that?)
• Iterative Methods by Space Decomposition and Subspace Correction
• The method of alternating projections and the method of subspace corrections in Hilbert space
• Iterative solution of large sparse systems of equations. Transl. from the German
• Elliptic differential equations: theory and numerical treatment. Transl. from the German by Regine Fadiman and Patrick D. F. Ion
• Title not available (Why is that?)
• Parallel Multilevel Preconditioners
• Algorithm-Based Fault Tolerance for Matrix Operations
• A Parallel Implementation of an Iterative Substructuring Algorithm for Problems in Three Dimensions
• Recovery Patterns for Iterative Methods in a Parallel Unstable Environment
• Title not available (Why is that?)
• Exaflop/s: the why and the how
• Title not available (Why is that?)
• Title not available (Why is that?)
• Finite Element Methods in Mechanics
• Algorithmic Fault Tolerance Using the Lanczos Method
• Numerical analysis of fixed point algorithms in the presence of hardware faults
• Title not available (Why is that?)

Cited In (9)

• Stochastic subspace correction methods and fault tolerance
• Algorithm-based error-detection schemes for iterative solution of partial differential equations
• Is the multigrid method fault tolerant? the two-grid case
• Modifying the asynchronous Jacobi method for data corruption resilience
• Efficient measurement error correction with spatially misaligned data
• Resilience for massively parallel multigrid solvers
• The resiliency of multilevel methods on next-generation computing platforms: probabilistic model and its analysis
• Randomized and fault-tolerant method of subspace corrections
• ROBUST SUBSPACE CORRECTION METHODS FOR NEARLY SINGULAR SYSTEMS

This page was built for publication: An error-resilient redundant subspace correction method

Report a bug (only for logged in users!)Click here to report a bug for this page (MaRDI item Q2359624)