[petsc-users] profiling question

[Jed Brown]

On Tue, Sep 21, 2010 at 13:49, Leo van Kampenhout
<lvankampenhout at gmail.com> wrote:
> Thanks for the helpful response Jed. I was not aware of the possibility to
> run seperate PETSC_COMM_WORLDS in the same program,  at least this is not
> clear from the documentation

This is really an MPI thing, you are using a different
PETSC_COMM_WORLD in different processes.  The way it works is that you
call MPI_Init, then MPI_Comm_split or whatever to create new
communicators for each group of processes, then

  PETSC_COMM_WORLD = new_comm;
  PetscInitialize(...);

The instances of PETSc with different PETSC_COMM_WORLDs are entirely
independent.

> About presenting scaling results, the most appealing to me seems to show two
> graphs, one with intra-node scaling (1-12) and the other going upwards from
> there (12, 24, 36, ...)

Yes, this is good.

Jed