On Wed, 16 Aug 2017, Renato Poli wrote:
I added this call on the master process (id=0): vector<double> _sys_solution; /// in .hxx sys.solution->localize_to_one( _sys_solution ); The system got stuck,
As I'd expect. The solution only gets filled on proc_id=0, but it's still a collective operation: proc 0 is the only one receiving, but every other processor is sending.
with all four processes with 100% processor usage.
That's MPI busy-waiting. In real use cases, where you have at least one core for every MPI rank and you're trying to run as fast as possible, I guess it's the lowest latency thing to do, but I wish I know how to turn it off and make mpich2 or openmpi use blocking waits instead. Sometimes I have one processor stopped by gdb and the others are just wasting CPU, sometimes I'd like to use N cores to run N*2 MPI ranks for debugging purposes and I'd like them not to step on each other's toes... --- Roy ------------------------------------------------------------------------------ Check out the vibrant tech community on one of the world's most engaging tech sites, Slashdot.org! http://sdm.link/slashdot _______________________________________________ Libmesh-users mailing list Libmesh-users@lists.sourceforge.net https://lists.sourceforge.net/lists/listinfo/libmesh-users
