[dl6tud]

I would like to read the meshes of all processors when computing in parallel. The idea is to have all meshes on each processor.

I create a "Time-Object" for each processor as

PtrList<Time> time(Pstream::nProcs());

forAll(time, procI)
{
time.set
(
procI,
new Time
(
Time::controlDictName,
args.rootPath(),
args.globalCaseName()/fileName(word("processor") + name(procI))
)
);
}

Then, I create the meshes with

PtrList<fvMesh> pMesh(Pstream::nProcs());
forAll(pMesh, procI)
{
pMesh.set
(
procI,
new fvMesh
(
IOobject
(
Foam::fvMesh::defaultRegion,
time[procI].timeName(),
time[procI],
IOobject::MUST_READ
)
)
);
}

The second step fails with

Message from processor 0 not fully consumed. messageSize:20620 bytes of which only 0 consumed.

At the moment I have no idea, why OF tries to communicate with the other processors when reading the meshes. Maybe for the boundaries?

[ngj]

Hi,

Yes, the processor boundaries exchange information, so what you are trying to do will most probably not work.

May I ask, why you need all the meshes on every processor?

Kind regards,

Niels

[A_Pete]

Hi,

what about using gather and scatter? Not sure what you exactly want to do with the mesh, but you could use gatherList.

//- Get local mesh on processor
fvMesh mesh(refCast<const fvMesh>(obr_));

//- Gather the meshes into PtrList on master processor
PtrList<fvMesh> pMesh(Pstream::nProcs());
pMesh[Pstream::myProcNo()] = mesh;
Pstream::gatherList(pMesh);

Depending on what you want to do afterwards, you could either use scatterList or put the mesh together again from the pMesh and then scatter just this fvMesh.