openfoam/applications/utilities/parallelProcessing/redistributePar/loadOrCreateMesh.C
Andrew Heather d8d6030ab6 INT: Integration of Mattijs' collocated parallel IO additions
Original commit message:
------------------------

Parallel IO: New collated file format

When an OpenFOAM simulation runs in parallel, the data for decomposed fields and
mesh(es) has historically been stored in multiple files within separate
directories for each processor.  Processor directories are named 'processorN',
where N is the processor number.

This commit introduces an alternative "collated" file format where the data for
each decomposed field (and mesh) is collated into a single file, which is
written and read on the master processor.  The files are stored in a single
directory named 'processors'.

The new format produces significantly fewer files - one per field, instead of N
per field.  For large parallel cases, this avoids the restriction on the number
of open files imposed by the operating system limits.

The file writing can be threaded allowing the simulation to continue running
while the data is being written to file.  NFS (Network File System) is not
needed when using the the collated format and additionally, there is an option
to run without NFS with the original uncollated approach, known as
"masterUncollated".

The controls for the file handling are in the OptimisationSwitches of
etc/controlDict:

OptimisationSwitches
{
    ...

    //- Parallel IO file handler
    //  uncollated (default), collated or masterUncollated
    fileHandler uncollated;

    //- collated: thread buffer size for queued file writes.
    //  If set to 0 or not sufficient for the file size threading is not used.
    //  Default: 2e9
    maxThreadFileBufferSize 2e9;

    //- masterUncollated: non-blocking buffer size.
    //  If the file exceeds this buffer size scheduled transfer is used.
    //  Default: 2e9
    maxMasterFileBufferSize 2e9;
}

When using the collated file handling, memory is allocated for the data in the
thread.  maxThreadFileBufferSize sets the maximum size of memory in bytes that
is allocated.  If the data exceeds this size, the write does not use threading.

When using the masterUncollated file handling, non-blocking MPI communication
requires a sufficiently large memory buffer on the master node.
maxMasterFileBufferSize sets the maximum size in bytes of the buffer.  If the
data exceeds this size, the system uses scheduled communication.

The installation defaults for the fileHandler choice, maxThreadFileBufferSize
and maxMasterFileBufferSize (set in etc/controlDict) can be over-ridden within
the case controlDict file, like other parameters.  Additionally the fileHandler
can be set by:
- the "-fileHandler" command line argument;
- a FOAM_FILEHANDLER environment variable.

A foamFormatConvert utility allows users to convert files between the collated
and uncollated formats, e.g.
    mpirun -np 2 foamFormatConvert -parallel -fileHandler uncollated

An example case demonstrating the file handling methods is provided in:
$FOAM_TUTORIALS/IO/fileHandling

The work was undertaken by Mattijs Janssens, in collaboration with Henry Weller.
2017-07-07 11:39:56 +01:00

411 lines
11 KiB
C

/*---------------------------------------------------------------------------*\
========= |
\\ / F ield | OpenFOAM: The Open Source CFD Toolbox
\\ / O peration |
\\ / A nd | Copyright (C) 2012-2017 OpenFOAM Foundation
\\/ M anipulation | Copyright (C) 2015 OpenCFD Ltd.
-------------------------------------------------------------------------------
License
This file is part of OpenFOAM.
OpenFOAM is free software: you can redistribute it and/or modify it
under the terms of the GNU General Public License as published by
the Free Software Foundation, either version 3 of the License, or
(at your option) any later version.
OpenFOAM is distributed in the hope that it will be useful, but WITHOUT
ANY WARRANTY; without even the implied warranty of MERCHANTABILITY or
FITNESS FOR A PARTICULAR PURPOSE. See the GNU General Public License
for more details.
You should have received a copy of the GNU General Public License
along with OpenFOAM. If not, see <http://www.gnu.org/licenses/>.
\*---------------------------------------------------------------------------*/
#include "loadOrCreateMesh.H"
#include "processorPolyPatch.H"
#include "processorCyclicPolyPatch.H"
#include "Time.H"
//#include "IOPtrList.H"
#include "polyBoundaryMeshEntries.H"
// * * * * * * * * * * * * * * * Global Functions * * * * * * * * * * * * * //
//namespace Foam
//{
// defineTemplateTypeNameAndDebug(IOPtrList<entry>, 0);
//}
// * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * //
// Read mesh if available. Otherwise create empty mesh with same non-proc
// patches as proc0 mesh. Requires all processors to have all patches
// (and in same order).
Foam::autoPtr<Foam::fvMesh> Foam::loadOrCreateMesh
(
const IOobject& io
)
{
// Region name
// ~~~~~~~~~~~
fileName meshSubDir;
if (io.name() == polyMesh::defaultRegion)
{
meshSubDir = polyMesh::meshSubDir;
}
else
{
meshSubDir = io.name()/polyMesh::meshSubDir;
}
// Patch types
// ~~~~~~~~~~~
// Read and scatter master patches (without reading master mesh!)
PtrList<entry> patchEntries;
if (Pstream::master())
{
//// Read PtrList of dictionary as dictionary.
//const word oldTypeName = IOPtrList<entry>::typeName;
//const_cast<word&>(IOPtrList<entry>::typeName) = word::null;
//IOPtrList<entry> dictList
//(
// IOobject
// (
// "boundary",
// io.time().findInstance
// (
// meshSubDir,
// "boundary",
// IOobject::MUST_READ
// ),
// meshSubDir,
// io.db(),
// IOobject::MUST_READ,
// IOobject::NO_WRITE,
// false
// )
//);
//const_cast<word&>(IOPtrList<entry>::typeName) = oldTypeName;
//// Fake type back to what was in field
//const_cast<word&>(dictList.type()) = dictList.headerClassName();
//
//patchEntries.transfer(dictList);
const fileName facesInstance = io.time().findInstance
(
meshSubDir,
"faces",
IOobject::MUST_READ
);
patchEntries = polyBoundaryMeshEntries
(
IOobject
(
"boundary",
facesInstance,
meshSubDir,
io.db(),
IOobject::MUST_READ,
IOobject::NO_WRITE,
false
)
);
// Send patches
for
(
int slave=Pstream::firstSlave();
slave<=Pstream::lastSlave();
slave++
)
{
OPstream toSlave(Pstream::commsTypes::scheduled, slave);
toSlave << patchEntries;
}
}
else
{
// Receive patches
IPstream fromMaster
(
Pstream::commsTypes::scheduled,
Pstream::masterNo()
);
fromMaster >> patchEntries;
}
// Dummy meshes
// ~~~~~~~~~~~~
// Check who has a mesh
const bool haveMesh = fileHandler().isFile
(
fileHandler().filePath
(
io.time().path()/io.instance()/meshSubDir/"faces"
)
);
if (!haveMesh)
{
bool oldParRun = Pstream::parRun();
Pstream::parRun() = false;
// Create dummy mesh. Only used on procs that don't have mesh.
IOobject noReadIO(io);
noReadIO.readOpt() = IOobject::NO_READ;
fvMesh dummyMesh
(
noReadIO,
xferCopy(pointField()),
xferCopy(faceList()),
xferCopy(labelList()),
xferCopy(labelList()),
false
);
// Add patches
List<polyPatch*> patches(patchEntries.size());
label nPatches = 0;
forAll(patchEntries, patchi)
{
const entry& e = patchEntries[patchi];
const word type(e.dict().lookup("type"));
const word& name = e.keyword();
if
(
type != processorPolyPatch::typeName
&& type != processorCyclicPolyPatch::typeName
)
{
dictionary patchDict(e.dict());
patchDict.set("nFaces", 0);
patchDict.set("startFace", 0);
patches[patchi] = polyPatch::New
(
name,
patchDict,
nPatches++,
dummyMesh.boundaryMesh()
).ptr();
}
}
patches.setSize(nPatches);
dummyMesh.addFvPatches(patches, false); // no parallel comms
// Add some dummy zones so upon reading it does not read them
// from the undecomposed case. Should be done as extra argument to
// regIOobject::readStream?
List<pointZone*> pz
(
1,
new pointZone
(
"dummyPointZone",
labelList(0),
0,
dummyMesh.pointZones()
)
);
List<faceZone*> fz
(
1,
new faceZone
(
"dummyFaceZone",
labelList(0),
boolList(0),
0,
dummyMesh.faceZones()
)
);
List<cellZone*> cz
(
1,
new cellZone
(
"dummyCellZone",
labelList(0),
0,
dummyMesh.cellZones()
)
);
dummyMesh.addZones(pz, fz, cz);
dummyMesh.pointZones().clear();
dummyMesh.faceZones().clear();
dummyMesh.cellZones().clear();
//Pout<< "Writing dummy mesh to " << dummyMesh.polyMesh::objectPath()
// << endl;
dummyMesh.write();
Pstream::parRun() = oldParRun;
}
// Read mesh
// ~~~~~~~~~
// Now all processors have a (possibly zero size) mesh so read in
// parallel
//Pout<< "Reading mesh from " << io.objectPath() << endl;
autoPtr<fvMesh> meshPtr(new fvMesh(io));
fvMesh& mesh = meshPtr();
// Sync patches
// ~~~~~~~~~~~~
if (!Pstream::master() && haveMesh)
{
// Check master names against mine
const polyBoundaryMesh& patches = mesh.boundaryMesh();
forAll(patchEntries, patchi)
{
const entry& e = patchEntries[patchi];
const word type(e.dict().lookup("type"));
const word& name = e.keyword();
if (type == processorPolyPatch::typeName)
{
break;
}
if (patchi >= patches.size())
{
FatalErrorInFunction
<< "Non-processor patches not synchronised."
<< endl
<< "Processor " << Pstream::myProcNo()
<< " has only " << patches.size()
<< " patches, master has "
<< patchi
<< exit(FatalError);
}
if
(
type != patches[patchi].type()
|| name != patches[patchi].name()
)
{
FatalErrorInFunction
<< "Non-processor patches not synchronised."
<< endl
<< "Master patch " << patchi
<< " name:" << type
<< " type:" << type << endl
<< "Processor " << Pstream::myProcNo()
<< " patch " << patchi
<< " has name:" << patches[patchi].name()
<< " type:" << patches[patchi].type()
<< exit(FatalError);
}
}
}
// Determine zones
// ~~~~~~~~~~~~~~~
wordList pointZoneNames(mesh.pointZones().names());
Pstream::scatter(pointZoneNames);
wordList faceZoneNames(mesh.faceZones().names());
Pstream::scatter(faceZoneNames);
wordList cellZoneNames(mesh.cellZones().names());
Pstream::scatter(cellZoneNames);
if (!haveMesh)
{
// Add the zones. Make sure to remove the old dummy ones first
mesh.pointZones().clear();
mesh.faceZones().clear();
mesh.cellZones().clear();
List<pointZone*> pz(pointZoneNames.size());
forAll(pointZoneNames, i)
{
pz[i] = new pointZone
(
pointZoneNames[i],
labelList(0),
i,
mesh.pointZones()
);
}
List<faceZone*> fz(faceZoneNames.size());
forAll(faceZoneNames, i)
{
fz[i] = new faceZone
(
faceZoneNames[i],
labelList(0),
boolList(0),
i,
mesh.faceZones()
);
}
List<cellZone*> cz(cellZoneNames.size());
forAll(cellZoneNames, i)
{
cz[i] = new cellZone
(
cellZoneNames[i],
labelList(0),
i,
mesh.cellZones()
);
}
mesh.addZones(pz, fz, cz);
}
// if (!haveMesh)
// {
// // We created a dummy mesh file above. Delete it.
// const fileName meshFiles = io.time().path()/io.instance()/meshSubDir;
// //Pout<< "Removing dummy mesh " << meshFiles << endl;
// mesh.removeFiles();
// rmDir(meshFiles);
// }
//
// Force recreation of globalMeshData.
// mesh.clearOut();
mesh.globalData();
// Do some checks.
// Check if the boundary definition is unique
mesh.boundaryMesh().checkDefinition(true);
// Check if the boundary processor patches are correct
mesh.boundaryMesh().checkParallelSync(true);
// Check names of zones are equal
mesh.cellZones().checkDefinition(true);
mesh.cellZones().checkParallelSync(true);
mesh.faceZones().checkDefinition(true);
mesh.faceZones().checkParallelSync(true);
mesh.pointZones().checkDefinition(true);
mesh.pointZones().checkParallelSync(true);
return meshPtr;
}
// ************************************************************************* //