[petsc-dev] Size of " PetscInt *garray" in Mat_MPIAIJ

[Fande Kong]

Does mat->cmap->N= mat->A->cmap->N + mat->B->cmap->N??

I maybe misunderstand this.

Fande,


On Thu, Mar 14, 2019 at 4:09 PM Jed Brown <jed at jedbrown.org> wrote:

> Fande Kong via petsc-dev <petsc-dev at mcs.anl.gov> writes:
>
> > Hi Developers,
> >
> > What is the size of " PetscInt *garray" in Mat_MPIAIJ?  Will it be
> > mat->B->cmap->N?
>
> Yes.
>
> > If so, it is not scalable in terms of the problem size, right? I am
> > missing something here.
>
> No, B only has columns for off-process entries where there are nonzeros.
> It's only a scalability issue if you have dense rows, in which case you
> have bigger problems on communication and should not use 1D matrix
> decompositions.
>
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://lists.mcs.anl.gov/pipermail/petsc-dev/attachments/20190314/ce56532e/attachment.html>