Numerical Analysis and Scientific Computing Seminar

On Nonsingular Saddle-Point Systems with a Maximally Rank-Deficient Leading Block

Speaker: Chen Greif, UBC

Location: Warren Weaver Hall 1302

Date: Nov. 20, 2015, 10 a.m.

Synopsis:

We consider nonsingular saddle-point matrices whose leading block is maximally rank deficient, and show that the inverse in this case has unique mathematical properties. We then develop a class of indefinite block preconditioners that rely on approximating the null space of the leading block. The preconditioned matrix is a product of two indefinite matrices, but under certain conditions the conjugate gradient method can be applied and is rapidly convergent. Spectral properties of the preconditioners are observed and validated by numerical experiments.

This is joint work with Ron Estrin.