Re: [petsc-dev] MatComputeExplicitOperator

[Smith, Barry F. via petsc-dev]

> On Apr 20, 2019, at 1:30 PM, Stefano Zampini <stefano.zamp...@gmail.com> 
> wrote:
> 
> Using preallocator will require looping twice for the computation of the 
> entries,

   No, that is totally impractical. The old code that Jed removed just 
preallocated the AIJ as dense and this was fine for this purpose

1) MatConvert_Shell() should call all the preallocating possibilities (treating 
as dense)
2) MatComputeExplicitOperator() could get a new second argument that is a 
matrix type (then we don't need to argue 
    if it should use DENSE or AIJ).



   Barry

> and this may be very expensive.
> In my case, every MatMult requires one forward and one backward TS solve.


> 
> We can either use AIJ or DENSE format, but I think this should be the same 
> for the sequential or parallel runs, not as it is now.
> 
> Il giorno sab 20 apr 2019 alle ore 14:20 Matthew Knepley <knep...@gmail.com> 
> ha scritto:
> On Fri, Apr 19, 2019 at 5:58 PM Smith, Barry F. via petsc-dev 
> <petsc-dev@mcs.anl.gov> wrote:
> 
> 
>    I think MPIAIJ was selected because it provided the most parallel 
> functionality compared to MPIDENSE for which more operations were not written.
>    This may not be relevant any more. 
> 
>     Definitely the code needs to be fixed. Fixing MatConvert_Shell(); it 
> should just assume 
> the matrices are dense and preallocate for them.
> 
> We should restore preallocation. It should be easy to do with Preallocator.
> 
>   Matt
>  
>    Barry
> 
> > On Apr 19, 2019, at 3:07 PM, Stefano Zampini via petsc-dev 
> > <petsc-dev@mcs.anl.gov> wrote:
> > 
> > What is the rationale behind MatComputeExplicitOperator returning SEQDENSE 
> > in sequential and MPIAIJ in parallel?
> > 
> > Also, before commit 
> > https://bitbucket.org/petsc/petsc/commits/b3d09e869df0e6ebcb615ca876706bfed4fcf1cd
> >   full preallocation of the MPIAIJ matrix was happening. Now, if we have a 
> > very dense operator that we want to sample just for testing purposes 
> > (inspecting entries, etc..) we have to pay the price of reallocating over 
> > and over.
> > 
> > What is the proper fix? 1) Use MPIDENSE? 2) Restore full preallocation for 
> > MPIAIJ? 3) Have MatComputeExplicitOperator to accept more arguments ? 
> > 
> > I'm in favor of 1
> > 
> > -- 
> > Stefano
> 
> 
> 
> -- 
> 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
> 
> https://www.cse.buffalo.edu/~knepley/
> 
> 
> -- 
> Stefano