>
>
> Nonetheless, algebraic multigrid seems to work really well compared to
> other preconditioners. Can I expect any further runtime improvements by
> using additional options like -pc_mg_smoothdown <1> or are the standard
> values generally sufficient?
>
Getting almost one digit reduction in re
For clarification (as it seems I may have misused certain terms):
Poisson's equation is meant to be solved ~1E6 times with successively
changing right handed side. Assembling the matrix wont be a dominating
part of the solution process as long as it can be reused. My initial
concerns involved p
On 3 June 2016 at 15:34, Michael Becker <
michael.bec...@physik.uni-giessen.de> wrote:
> So using -log_summary helps me find out how much time is actually spent on
> the PETSc routines that are repeatedly called. Since that part of my code
> is fairly simple:
>
> PetscScalar *barray;
> VecGetA
On Fri, Jun 3, 2016 at 9:34 AM, Michael Becker <
michael.bec...@physik.uni-giessen.de> wrote:
> So using -log_summary helps me find out how much time is actually spent on
> the PETSc routines that are repeatedly called. Since that part of my code
> is fairly simple:
>
> PetscScalar *barray;
>
So using -log_summary helps me find out how much time is actually spent
on the PETSc routines that are repeatedly called. Since that part of my
code is fairly simple:
PetscScalar *barray;
VecGetArray(b,&barray);
for (int i=0; i, I don't see how additional log states would help me. So I wo
On Fri, Jun 3, 2016 at 5:56 AM, Dave May wrote:
> On 3 June 2016 at 11:37, Michael Becker <
> michael.bec...@physik.uni-giessen.de> wrote:
>
>> Dear all,
>>
>> I have a few questions regarding possible performance enhancements for
>> the PETSc solver I included in my project.
>>
>> It's a particl
On 3 June 2016 at 11:37, Michael Becker <
michael.bec...@physik.uni-giessen.de> wrote:
> Dear all,
>
> I have a few questions regarding possible performance enhancements for the
> PETSc solver I included in my project.
>
> It's a particle-in-cell plasma simulation written in C++, where Poisson's
>
Dear all,
I have a few questions regarding possible performance enhancements for
the PETSc solver I included in my project.
It's a particle-in-cell plasma simulation written in C++, where
Poisson's equation needs to be solved repeatedly on every timestep.
The simulation domain is discretized
