Parallel Factorizations in Numerical Analysis
نویسندگان
چکیده
In this paper we review the parallel solution of sparse linear systems, usually deriving by the discretization of ODEIVPs or ODE-BVPs. The approach is based on the concept of parallel factorization of a (block) tridiagonal matrix. This allows to obtain efficient parallel extensions of many known matrix factorizations, and to derive, as a by-product, a unifying approach to the parallel solution of ODEs.
منابع مشابه
Parallel Incomplete Cholesky Preconditioners Based on the Non-Overlapping Data Distribution
The paper analyses various parallel incomplete factorizations based on the non-overlapping domain decomposition. The general framework is applied to the investigation of the preconditioning step in cg-like methods. Under certain conditions imposed on the nite element mesh, all matrix and vector types given by the special data distribution can be used in the matrix-by-vector multiplications. Not...
متن کاملSpectral analysis of parallel incomplete factorizations with implicit pseudo-overlap
Two general parallel incomplete factorization strategies are investigated. The techniques may be interpreted as local re-numberings of nodes, combined with suitable ll level in a (successful) attempt to overcome the well-known trade-oo between high parallelism (locality) and fast convergence (globality). Spectral analysis is provided, which displays how the condition number depends on the numbe...
متن کاملParallel Band Two-Sided Matrix Bidiagonalization for Multicore Architectures
The objective of this paper is to extend, in the context of multicore architectures, the concepts of algorithms-by-tiles [Buttari et al., 2007] for Cholesky, LU, QR factorizations to the family of twosided factorizations. In particular, the bidiagonal reduction of a general, dense matrix is very often used as a pre-processing step for calculating the singular value decomposition. Furthermore, i...
متن کاملApplying Parallel Direct Solver Techniques to Build Robust High Performance Preconditioners
The purpose of our work is to provide a method which exploits the parallel blockwise algorithmic approach used in the framework of high performance sparse direct solvers in order to develop robust preconditioners based on a parallel incomplete factorization. The idea is then to define an adaptive blockwise incomplete factorization that is much more accurate (and numerically more robust) than th...
متن کاملA Numerical Study of Some Parallel Algebraic Preconditioners
We present a numerical study of several parallel algebraic preconditioners, which speed up the convergence of Krylov iterative methods when solving large-scale linear systems. The studied algebraic preconditioners are of two types. The first type includes simple block preconditioners using incomplete factorizations or preconditioned Krylov solvers on the subdomains. The second type is an enhanc...
متن کاملذخیره در منابع من
با ذخیره ی این منبع در منابع من، دسترسی به آن را برای استفاده های بعدی آسان تر کنید
عنوان ژورنال:
- Scalable Computing: Practice and Experience
دوره 10 شماره
صفحات -
تاریخ انتشار 2009