A Robust Algebraic Domain Decomposition Preconditioner for Sparse Normal Equations

نویسندگان

چکیده

Solving the normal equations corresponding to large sparse linear least-squares problems is an important and challenging problem. For very problems, iterative solver needed and, in general, a preconditioner required achieve good convergence. In recent years, number of preconditioners have been proposed. These are largely serial, reported results demonstrate that none commonly used for matrix capable solving all problems. Our interest thus designing new efficient robust can be implemented parallel. proposed constructed efficiently algebraically without any knowledge problem assumption on except it sparse. We exploit structure symmetric positive definite use concept algebraic local semidefinite splittings introduce two-level Schwarz The condition preconditioned shown theoretically bounded independently subdomains splitting. This upper bound adjusted using single parameter $\tau$ user specify. discuss how top PETSc library only 150 lines Fortran, C, or Python code. Problems arising from practical applications compare performance with other preconditioners.

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

اگر عضو سایت هستید لطفا وارد حساب کاربری خود شوید

منابع مشابه

A Robust Algebraic Preconditioner for Nite Diierence Approximations of Convection{diiusion Equations

Stable nite di erence approximations of convection-di usion equations lead to large sparse linear systems of equations whose coe cient matrix is an M{matrix, which is highly non symmetric when the convection dominates. For an e cient iterative solution of such systems, it is proposed to consider in the non symmetric case an algebraic multilevel preconditioning method formerly proposed for pure ...

متن کامل

A robust algebraic multilevel preconditioner for non-symmetric M-matrices

Stable finite difference approximations of convection–diffusion equations lead to large sparse linear systems of equations whose coefficient matrix is an M-matrix, which is highly non-symmetric when the convection dominates. For an efficient iterative solution of such systems, it is proposed to consider in the non-symmetric case an algebraic multilevel preconditioning method formerly proposed f...

متن کامل

Parallel Subdomain-based Preconditioner for Non-overlapping Domain Decomposition Methods Parallel Subdomain-based Preconditioner for Non-overlapping Domain Decomposition Methods

We present a new parallelizable preconditioner that is used as the local component for a two-level preconditioner similar to BPS. On 2D model problems that exhibit either high anisotropy or discontinuity, we demonstrate its attracting numerical behaviour and compare it to the regular BPS. Finally, to alleviate the construction cost of this new preconditioner, that requires the explicit computat...

متن کامل

A domain decomposition preconditioner for multiscale high-contrast problems

We present a new class of coarse spaces for two-level additive Schwarz preconditioners that yield condition number bound independent of the contrast in the media properties. These coarse spaces are an extension of the spaces discussed in [3]. Second order elliptic equations are considered. We present theoretical and numerical results. Detailed description of the results and numerical studies wi...

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

ژورنال

عنوان ژورنال: SIAM Journal on Scientific Computing

سال: 2022

ISSN: ['1095-7197', '1064-8275']

DOI: https://doi.org/10.1137/21m1434891