A parallel implementation of the modified augmented Lagrangian preconditioner for the incompressible Navier-Stokes equations

We describe a parallel implementation of a block triangular preconditioner based on the modified augmented Lagrangian approach to the steady incompressible Navier-Stokes equations. The equations are linearized by Picard iteration and discretized with various finite element and finite difference schemes on two- and three-dimensional domains. We report strong scalability results for up to 64 cores. © 2012 Springer Science+Business Media New York