An Approach for Parallelizing any General Unsymmetric Sparse Matrix Algorithm
نویسندگان
چکیده
In many large scale scientiic and engineering computations, the solution to a sparse linear system is required. We present a partial unsymmetric nested dissection method that can be used to parallelize any general unsymmetric sparse matrix algorithm whose pivot search can be restricted to a subset of rows and columns in the active submatrix. The submatrix is determined by the partial unsymmetric dissection. We present results of this approach, applied to the unsymmetric-pattern multifrontal method.
منابع مشابه
A Novel Parallel Algorithm for Gaussian Elimination of Sparse Unsymmetric Matrices
We describe a new algorithm for Gaussian Elimination suitable for general (unsymmetric and possibly singular) sparse matrices of any entry type, which has a natural parallel and distributed-memory formulation but degrades gracefully to sequential execution. We present a sample MPI implementation of a program computing the rank of a sparse integer matrix using the proposed algorithm. Some prelim...
متن کاملQuantitative Methods Inquires 179 IMPROVING THE PERFORMANCE OF SPARSE LU MATRIX FACTORIZATION USING A SUPERNODAL ALGORITHM
In this paper we investigate a method to improve the performance of sparse LU matrix factorization used to solve unsymmetric linear systems, which appear in many mathematical models. We introduced and used the concept of the supernode for unsymmetric matrices in order to use dense matrix operations to perform the LU factorization for sparse matrices. We describe an algorithm that uses supernode...
متن کاملApplying Schur Complements for Handling General Updates of a Large, Sparse, Unsymmetric Matrix
We describe a set of procedures for computing and updating an inverse representation of a large and sparse unsymmetric matrix A. The representation is built of two matrices: an easily invertible, large and sparse matrix A0 and a dense Schur complement matrix S. An e cient heuristic is given that nds this representation for any matrix A and keeps the size of S as small as possible. Equations wit...
متن کاملAn Unsymmetric-Pattern Multifrontal Method for Sparse LU Factorization
Sparse matrix factorization algorithms for general problems are typically characterized by irregular memory access patterns that limit their performance on parallel-vector supercomputers. For symmetric problems, methods such as the multifrontal method avoid indirect addressing in the innermost loops by using dense matrix kernels. However, no efficient LU factorization algorithm based primarily ...
متن کاملA CPU-GPU hybrid approach for the unsymmetric multifrontal method
Multifrontal is an efficient direct method for solving large-scale sparse and unsymmetric linear systems. The method transforms a large sparse matrix factorization process into a sequence of factorizations involving smaller dense frontal matrices. Some of these dense operations can be accelerated by using a graphic processing unit (GPU). We analyze the unsymmetricmultifrontalmethod fromboth an ...
متن کامل