act a (minor) bug in open-mpi 1.8 that would be
fixed in the 1.8.3 release. So it seems there's no real need to
prolong the discussion of this point.
Allin Cottrell
On Sat, 6 Sep 2014, Ralph Castain wrote:
On Sep 6, 2014, at 7:52 AM, Allin Cottrell wrote:
On Fri, 5 Sep 2014, Ralph Castain wrote:
On Sep 5, 2014, at 3:34 PM, Allin Cottrell wrote:
I suspect there is a new (to openmpi 1.8.N?) warning with respect to
requesting a number of MPI processes
On Fri, 5 Sep 2014, Ralph Castain wrote:
On Sep 5, 2014, at 3:34 PM, Allin Cottrell wrote:
I suspect there is a new (to openmpi 1.8.N?) warning with respect to
requesting a number of MPI processes greater than the number of "real"
cores on a given machine. I can provide a good
er friendly) warning whenever I specify to mpiexec a
number of processes > the number of "real" cores -- even if the
computation is in fact completed faster that way -- and I'd like to
track down what has changed.
As I indicated, I'll be happy to provide more inform
