Dynamic Pivoting by Shcur Decomposition Based Method for GLU Solver
نویسندگان
چکیده
This document presents the Schur-decomposition based dynamic pivoting method for GLU solver. The new pivoting method allows dynamic pivoting for small number of small valued pivots after the AMD (approximate minimum degree) ordering process and the symbolic analysis process in the GLU, which is similar to the left-looking method. This method will not be very effective if there is a large numbers of pivoting as each pivoting requires adding one row and column. Section 1 explains the Schur complement and a pivoting solution method for a particular example. Section 2 shows how to eliminate small pivots and put the original system in the format of section 1. Section 3 gives a simple example. Section 4 presents an algorithm and discusses implementation issues.
منابع مشابه
A Randomized LU-based Solver Using GPU and Intel Xeon Phi Accelerators
We present a fast hybrid solver for dense linear systems based on LU factorization. To achieve good performance, we avoid pivoting by using random butterfly transformations for which we developed efficient implementations on heterogeneous architectures. We used both Graphics Processing Units and Intel Xeon Phi as accelerators. The performance results show that the pre-processing due to randomiz...
متن کاملHigh performance simplex solver
The dual simplex method is frequently the most efficient technique for solving linear programming (LP) problems. This thesis describes an efficient implementation of the sequential dual simplex method and the design and development of two parallel dual simplex solvers. In serial, many advanced techniques for the (dual) simplex method are implemented, including sparse LU factorization, hyper-spa...
متن کاملA Cholesky Algorithm for Some Complex Symmetric Systems
Complex symmetric systems do not in general admit a Cholesky factorization without pivoting, as would be the case for hermitian systems. Nevertheless, for some complex symmetric systems, as those coming from the discretization of boundary integral formulations, pivoting can be avoided. We present a Cholesky factorization algorithm for such complex symmetric systems. We propose a LAPACK-style im...
متن کاملBIT ? ? ( 199 ? ) , ? ? ? { ? ? ? . An Assessment of Incomplete - LU Preconditioners forNonsymmetric Linear Systems
We report on an extensive experiment to compare an iterative solver preconditioned by several versions of incomplete LU factorization with a sparse direct solver using LU factorization with partial pivoting. Our test suite is 24 nonsymmetric matrices drawn from benchmark sets in the literature. On a few matrices, the best iterative method is more than 5 times as fast and more than 10 times as m...
متن کاملA Scalable Sparse Direct Solver Using Static Pivoting
We propose several techniques as alternatives to partial pivoting to stabilize sparse Gaussian elimination. From numerical experiments we demonstrate that for a wide range of problems the new method is as stable as partial pivoting. The main advantage of the new method over partial pivoting is that it permits a priori determination of data structures and communication pattern, which makes it mo...
متن کامل