An efficient deflation technique for the communication-avoiding conjugate gradient

MaRDI QIDQ2341373zbMATH OpenFDO

Authors Nicholas Knight, James Demmil, Erin Carson

Publication date 24 April 2015

Published in ETNA - Electronic Transactions on Numerical Analysis (Search for Journal in Brave)

Full work available at URL http://www.emis.de/journals/ETNA/volumes/2011-2020/vol43/abstract_vol43_pp125-141.html

IPFS File on IPFS (Hint: this is only the Hash - if you get a timeout, this file is not available on our server.)

zbMATH Keywords

sparse matricesKrylov subspace methodsnumerical experimentpreconditionerconjugate gradient methoditerative solversdeflationminimizing communication

Mathematics Subject Classification ID

Computational methods for sparse matrices (65F50) Parallel numerical computation (65Y05) Complexity and performance of numerical algorithms (65Y20) Preconditioners for iterative methods (65F08) Iterative numerical methods for linear systems (65F10)

Recommendations

• Enlarged Krylov subspace conjugate gradient methods for reducing communication
• Avoiding communication in nonsymmetric Lanczos-based Krylov subspace methods
• Communication Avoiding ILU0 Preconditioner
• A residual replacement strategy for improving the maximum attainable accuracy of \(s\)-step Krylov subspace methods
• The adaptive \(s\)-step conjugate gradient method

Cited in

(4)

• Avoiding Communication in Primal and Dual Block Coordinate Descent Methods
• Avoiding communication in nonsymmetric Lanczos-based Krylov subspace methods
• s-step enlarged Krylov subspace conjugate gradient methods
• Enlarged Krylov subspace conjugate gradient methods for reducing communication

This page was built for publication: An efficient deflation technique for the communication-avoiding conjugate gradient

Report a bug (only for logged in users!)Click here to report a bug for this page (MaRDI item Q2341373)