PETSc runs slower on a shared memory machine than on a cluster

[Lisandro Dalcin]

On 2/2/07, Shi Jin <jinzishuai at yahoo.com> wrote:
> Thank you for replying.
> I run the code on a shared memory machine through MPI,
> just like what I do on a cluster. I simply did:
> petscmpirun -np 18 ./code
>
> I am not 100% sure whether MPICH-2 will automatically
> use shared memory as the underlying commnunication
> channel instead of the network but I know most MPI
> implementations are smart enough to do so (like
> LAM-MPI I used before). Could anyone confirm this?

Please read the following...

http://www-unix.mcs.anl.gov/mpi/mpich/downloads/mpich2-doc-README.txt

I think for shared-memory you should try to configure MPICH2 with the following:

--with-device=ch3:shm

If not, perhaps configure will default to --with-device=ch3:sock and
MPICH2 will use TCP sockets.

I hope this help you.

Regards,

-- 
Lisandro Dalcín
---------------
Centro Internacional de Métodos Computacionales en Ingeniería (CIMEC)
Instituto de Desarrollo Tecnológico para la Industria Química (INTEC)
Consejo Nacional de Investigaciones Científicas y Técnicas (CONICET)
PTLC - Güemes 3450, (3000) Santa Fe, Argentina
Tel/Fax: +54-(0)342-451.1594