[petsc-dev] [pflotran-dev: 976] Re: [pflotran-users: 998] Distorted structured grid

[Paolo Orsini]

MatConvert was slow when I tested 10 days ago, because was not
preallocating the new matrix (see initial part of this thread). I was
trying to find a work around.
Unless you already fixed this, did you?... I haven't reinstalled petsc from
then

I need the final Dual Graph (the filtered matrix) in Adj mat format.
I will extract indices from the AIJ matrix (the matrix to filter), and I I
will create a new Adj (avoiding to do an in-place operation as you
suggested)

Thanks

Paolo




On Fri, Nov 15, 2013 at 7:34 PM, Jed Brown <jedbrown at mcs.anl.gov> wrote:

> Paolo Orsini <paolo.orsini at gmail.com> writes:
>
> > Hi Jed,
> >
> > I am back to work on this now...
> >
> > Thanks for explaining how I can do this. it makes perfectly sense:
> >
> > A. Assign entries values = 1, when forming the Adj matrix,
> >
> > B. Convert the Adj matrix to AIJ matrix, so I can use MatMatTransposeMult
> > (or form AIJ directly): can I preallocate AIJ before calling MatConvert?
> To
> > avoid slow performance? Is there any conflict between the preallocoation
> > functions and MatConvert?
>
> MatConvert() should be fast.
>
> > C. Filter AIJ, to remove entries with values <3 (for 3d problems). What
> > function shall I use to do this operation?
>
> Don't try to do it in-place.  Do you want the result in a matrix or in
> another data structure?  Anyway, I would just walk along calling
> MatGetRow() and picking out the column indices that have values of at
> least 3.
>
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://lists.mcs.anl.gov/pipermail/petsc-dev/attachments/20131115/c05dced9/attachment.html>