On evaluating elimination tree based parallel sparse cholesky factorizations

From MaRDI portal





The computation of the Cholesky factorization of sparse matrices by parallel methods is investigated. On the basis of the elimination tree model, the four widely adopted parallel methods -- the column-, row-, submatrix-Cholesky-, and multifrontal method -- are compared with respect to the computational costs. The theoretical result that the multifrontal method is superior to the other three Cholesky factorization methods is apparently confirmed by the experimental results.





Describes a project that uses

Uses Software






This page was built for publication: On evaluating elimination tree based parallel sparse cholesky factorizations

Report a bug (only for logged in users!)Click here to report a bug for this page (MaRDI item Q4955154)