Many thanks for your encouragement.
I’m neither an expert on MPI nor on thread. I underestimated the
difficulty.
This may take a very long time for me to figure out.
Hopefully some experts on these parallel computing regimes would like to
address such a curious problem soon.
Thanks！
On 05/23/2017 07:24 PM, Jack wrote:
Originally, I suspected that when solving two linear systems simultaneously by
two threads would reduce time. But now it seems that this idea increases the
complexity of communication between MPI communicators and coding, and also not
surely to decrease compu
Hi Prof. Wolfgang,
Many thanks!
Originally, I suspected that when solving two linear systems simultaneously
by two threads would reduce time. But now it seems that this idea increases
the complexity of communication between MPI communicators and coding, and
also not surely to decrease
Jack,
“The way to do this is to clone the MPI communicator you use for your
overall
problem once for each linear system. ”
That means for my problem I have to copy the Vector and Matrix of one
linear system（either the thermal diffusion or stokes flow） to another
Vector and Matrix which are
Hi Prof. Wolfgang，
Thanks so much！
“The way to do this is to clone the MPI communicator you use for your
overall
problem once for each linear system. ”
That means for my problem I have to copy the Vector and Matrix of one
linear system（either the thermal diffusion or stokes flow） to anoth
Jack,
unrelated to the question about versions of MPI libraries, when you solve two
linear systems on separate threads, you need to pay attention that the two
solvers (and all associated objects such as matrices, vectors, etc) use
separate MPI communicators. Otherwise you will have the two sol
Hi Bruno,
I changed the corresponding code and remake the library again,
unfortunately, the same warn came out as before.
I changed this in another workstation which uses MPICH2, it also failed
during solving the two systems.
I'm trying to update my OpenMPI to v2.1, but in the latest version,
Jack,
2017-05-19 22:06 GMT-04:00 Jack :
> I should update my OpenMPI to the latest version and such problem will be
> avoided and I do not need to change my code for initialization of MPI and
> TBB myself?
Unfortunately, you wil need to initialize MPI yourself or you can
change this line
https://
Hi Bruno,
I appreciated it very much for your responses.
The Matrices and Vectors are of TrilinosWrappers, the solvers should use
MPI, too. I used the OpenMPI 1.8.1. The release date is Apr 22 2014, later
than the post on the Github. I initialized the MPI as following
try
{
us
Jack,
are your solvers using MPI? This looks similar to this problem
https://github.com/open-mpi/ompi/issues/1081 Which version of MPI are you
using? How do you initialize MPI? MPI_InitFinalize set
MPI_THREAD_SERIALIZED which "tells MPI that
we might use several threads but never call two MPI
10 matches
Mail list logo
