> On May 1, 2014, at 8:19 PM, Matthew Knepley <[email protected]> wrote: > >> On Thu, May 1, 2014 at 6:14 PM, Miguel Angel Salazar de Troya >> <[email protected]> wrote: >> Hello everybody >> >> I want to implement an adaptive mesh refinement library in a code written in >> petsc. I have checked out some of the available libraries, but I want to >> work with the latest petsc-dev version and I am sure there will be many >> incompatibilities. So far I think I'll end up working with one of these >> libraries: SAMRAI, Chombo, libMesh and deal II. Before I start checking out >> each of them and learn how to use them I though I would ask you guys which >> one you would recommend. My code would be a finite element analysis in solid >> mechanics. I would like to take full advantage of petsc capabilities, but I >> would not mind start with some restrictions. I hope my question is not too >> broad. > > SAMRAI, Chombo, and Deal II are all structured adaptive refinement codes, > whereas LibMesh is unstructured. If you want unstructured, there is > really no other game in town. If you use deal II, I would suggest trying out > p4est underneath which gives great scalability. My understanding > is that Chombo is mostly used for finite volume and SAMRAI and deal II for > finite element, but this could be out of date.
SAMRAI is definitely much better suited to finite volume, although it does have basic features needed for structured-grid FE. -- Boyce > > Matt > >> Take care >> Miguel >> >> -- >> Miguel Angel Salazar de Troya >> Graduate Research Assistant >> Department of Mechanical Science and Engineering >> University of Illinois at Urbana-Champaign >> (217) 550-2360 >> [email protected] > > > > -- > What most experimenters take for granted before they begin their experiments > is infinitely more interesting than any results to which their experiments > lead. > -- Norbert Wiener
