Block GMRES Method with Inexact Breakdowns and Deflated Restarting

نویسندگان

  • Emmanuel Agullo
  • Luc Giraud
  • Yan-Fei Jing
چکیده

We consider the solution of large linear systems with multiple right-hand sides using a block GMRES approach. We introduce a new algorithm that effectively handles the situation of almost rank deficient block generated by the block Arnoldi procedure and that enables the recycling of spectral information at restart. The first feature is inherited from an algorithm introduced by Robbé and Sadkane [M. Robbé and M. Sadkane. Exact and inexact breakdowns in the block GMRES method. Linear Algebra and its Applications, 419: 265-285, 2006.], while the second one is obtained by extending the deflated restarting strategy proposed by Morgan [R. B. Morgan. Restarted block GMRES with deflation of eigenvalues. Applied Numerical Mathematics, 54(2): 222-236, 2005.]. Through numerical experiments, we show that the new algorithm combines efficiently the attractive numerical features of its two parents that it outperforms. Key-words: Block GMRES, multiple right-hand sides, numerical rank deficiency, subspace augmentation, harmonic Ritz values ∗ Inria, CNRS (LaBRI UMR 5800) and Université de Bordeaux † School of Mathematical Sciences/Institute of Computational Science, University of Electronic Science and Technology of China, Chengdu, Sichuan, 611731, P. R. China ([email protected] or [email protected]). This author was supported by INRIA fund, NSFC (61170311, 11201055), Chinese Universities Specialized Research Fund for the Doctoral Program (20120185120026), the State Scholarship Fund, the Fundamental Research Funds for the Central Universities. GMRES par bloc avec inexact breakdowns et deflation au restart Résumé : Nous considérons la résolution de systèmes linéaires avec second-membres multiplques par une approche GMRES par bloc. Nous introduisons un nouvel algorithme qui gère efficacement d’une part la situation de perte de rang numérique dans les blocs générés par la méthode d’Arnoldi d’autre part le recyclage d’information spectrale au redemarrage via une technique d’augmentation d’espace. La première propriété est hérité de l’algorithme introduit par Robbé and Sadkane [M. Robbé and M. Sadkane. Exact and inexact breakdowns in the block GMRES method. Linear Algebra and its Applications, 419: 265-285, 2006.], et la seconde est obtenue en étendant la stratégie de redémmarage proposée par Morgan [R. B. Morgan. Restarted block GMRES with deflation of eigenvalues. Applied Numerical Mathematics, 54(2): 222-236, 2005.]. Via des expérimentations numériques, nous montrons que ce nouvel algorithme combine efficacement les deux propriétés de ces deux parents dont il améliore les performances. Mots-clés : bloc GMRES, second-membres multiples, perte de rang numérique, augmentation de sous-espace, valeurs de Ritz harmoniques Block GMRES-DR method with inexact breakdowns 3

برای دانلود متن کامل این مقاله و بیش از 32 میلیون مقاله دیگر ابتدا ثبت نام کنید

ثبت نام

اگر عضو سایت هستید لطفا وارد حساب کاربری خود شوید

منابع مشابه

GMRES with Deflated Restarting

A modification is given of the GMRES iterative method for nonsymmetric systems of linear equations. The new method deflates eigenvalues using Wu and Simon’s thick restarting approach. It has the efficiency of implicit restarting, but is simpler and does not have the same numerical concerns. The deflation of small eigenvalues can greatly improve the convergence of restarted GMRES. Also, it is de...

متن کامل

Restarted Block Gmres with Deflation of Eigenvalues

Block-GMRES is an iterative method for solving nonsymmetric systems of linear equations with multiple right-hand sides. Restarting may be needed, due to orthogonalization expense or limited storage. We discuss how restarting affects convergence and the role small eigenvalues play. Then a version of restarted block-GMRES that deflates eigenvalues is presented. It is demonstrated that deflation c...

متن کامل

A study on block flexible iterative solvers with applications to Earth imaging problem in geophysics

This work is concerned with the development and study of a minimum residual norm subspace method based on the Generalized Conjugate Residual method with inner Orthogonalization (GCRO) method that allows flexible preconditioning and deflated restarting for the solution of non-symmetric or non-Hermitian linear systems. First we recall the main features of Flexible Generalized Minimum Residual wit...

متن کامل

A Flexible Generalized Conjugate Residual Method with Inner Orthogonalization and Deflated Restarting

This work is concerned with the development and study of a minimum residual norm subspace method based on the generalized conjugate residual method with inner orthogonalization (GCRO) method that allows flexible preconditioning and deflated restarting for the solution of nonsymmetric or non-Hermitian linear systems. First we recall the main features of flexible generalized minimum residual with...

متن کامل

ذخیره در منابع من


  با ذخیره ی این منبع در منابع من، دسترسی به آن را برای استفاده های بعدی آسان تر کنید

برای دانلود متن کامل این مقاله و بیش از 32 میلیون مقاله دیگر ابتدا ثبت نام کنید

ثبت نام

اگر عضو سایت هستید لطفا وارد حساب کاربری خود شوید

عنوان ژورنال:
  • SIAM J. Matrix Analysis Applications

دوره 35  شماره 

صفحات  -

تاریخ انتشار 2014