[petsc-users] Unstructured mesh

[Mark Adams]

> It is not. The number of processes is specified independently using
> 'mpiexec -n <p>' or when using the test system NP=<p>.
>
>> (i)     Say I start with -cells 1,1,1 -max_conv its 7; that would
>> eventually leave all refinement on level 7 running on 1 MPI process?
>>
> I don't understand

> (ii)    Say I start with -cells 2,2,1 -max_conv its n; is it recommended
>> to run on 4 MPI processes?
>>
> Yes!


> No, those options do not influence the number of processes.
>

Correct, but you are not answering the question that he asked!

The coupling of the two is not enforced in the code, but to get good load
balancing you want i*j*k == NP. (there is more flexibility than this, but
this is the place to start.)


>
>
>> I am running ex56 on gpu; I am looking at KSPSolve (or any other event)
>> but no gpu flops are recorded in the -log_view?
>>
>
> I do not think you are running on the GPU then. Mark can comment, but we
> usually specify GPU execution using the Vec and Mat types
> through -dm_vec_type and -dm_mat_type.
>

This test has a ex56_ prefix.

Mark
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://lists.mcs.anl.gov/pipermail/petsc-users/attachments/20211213/315b2136/attachment.html>