On the GPGPU parallelization issues of finite element approximate inverse preconditioning
نویسندگان
چکیده
منابع مشابه
Parallel Finite Element Approximate Inverse Preconditioning on Symmetric Multiprocessor Systems
Parallel normalized preconditioned conjugate gradient type methods based on normalized approximate finite element inverse matrix techniques are investigated for the efficient solution of sparse linear systems. Application of the proposed methods on a three dimensional boundary value problems is discussed and numerical results are given. The parallel implementation of the normalized precondition...
متن کاملApproximate Inverse Preconditioning for Shifted Linear Systems
In this paper we consider the problem of preconditioning symmetric positive definite matrices of the form Aα = A+αI where α > 0. We discuss how to cheaply modify an existing sparse approximate inverse preconditioner for A in order to obtain a preconditioner for Aα. Numerical experiments illustrating the performance of the proposed approaches are presented. AMS subject classification (2000): 65F...
متن کاملOn Some New Developments in Approximate Inverse Preconditioning
Some new developments of conjugation-based sparse approximate inverse preconditioners are considered in application to problems of solid and structural mechanics. A block version of the AINV algorithm and its stabilized modification are presented. The stabilized AINV algorithm, which in exact arithmetic is breakdown-free for any SPD matrix, is found to result in robust preconditioners even in d...
متن کاملApproximate inverse preconditioning in the parallel solution of sparse eigenproblems
A preconditioned scheme for solving sparse symmetric eigenproblems is proposed. The solution strategy relies upon the DACG algorithm, which is a Preconditioned Conjugate Gradient algorithm for minimizing the Rayleigh Quotient. A comparison with the well established ARPACK code, shows that when a small number of the leftmost eigenpairs is to be computed, DACG is more efficient than ARPACK. Effec...
متن کاملPreconditioning with Direct Approximate Factoring of the Inverse
To precondition a large and sparse linear system, two direct methods for approximate factoring of the inverse are devised. The algorithms are fully parallelizable and appear to be more robust than the iterative methods suggested for the task. A method to compute one of the matrix subspaces optimally is derived. Possessing a considerable amount of flexibility, these approaches extend the approxi...
متن کاملذخیره 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: Journal of Computational and Applied Mathematics
سال: 2011
ISSN: 0377-0427
DOI: 10.1016/j.cam.2011.07.016