Files
OpenFOAM-5.x/applications/utilities/surface/surfaceMeshTriangulate/surfaceMeshTriangulate.C
Henry Weller 7c301dbff4 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

379 lines
11 KiB
C

/*---------------------------------------------------------------------------*\
========= |
\\ / F ield | OpenFOAM: The Open Source CFD Toolbox
\\ / O peration |
\\ / A nd | Copyright (C) 2011-2017 OpenFOAM Foundation
\\/ M anipulation |
-------------------------------------------------------------------------------
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/>.
Application
surfaceMeshTriangulate
Description
Extracts surface from a polyMesh. Depending on output surface format
triangulates faces.
Region numbers on faces cannot be guaranteed to be the same as the patch
indices.
Optionally only triangulates named patches.
If run in parallel the processor patches get filtered out by default and
the mesh gets merged (based on topology).
\*---------------------------------------------------------------------------*/
#include "MeshedSurface.H"
#include "UnsortedMeshedSurface.H"
#include "argList.H"
#include "Time.H"
#include "polyMesh.H"
#include "processorPolyPatch.H"
#include "ListListOps.H"
#include "uindirectPrimitivePatch.H"
#include "globalMeshData.H"
#include "globalIndex.H"
using namespace Foam;
// * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * //
int main(int argc, char *argv[])
{
argList::addNote
(
"extract surface from a polyMesh"
);
argList::validArgs.append("output surface file");
#include "addRegionOption.H"
argList::addBoolOption
(
"excludeProcPatches",
"exclude processor patches"
);
argList::addOption
(
"patches",
"(patch0 .. patchN)",
"only triangulate selected patches (wildcards supported)"
);
argList::addOption
(
"faceZones",
"(fz0 .. fzN)",
"triangulate selected faceZones (wildcards supported)"
);
#include "setRootCase.H"
#include "createTime.H"
const fileName outFileName(args[1]);
Info<< "Extracting surface from boundaryMesh ..."
<< endl << endl;
const bool includeProcPatches =
!(
args.optionFound("excludeProcPatches")
|| Pstream::parRun()
);
if (includeProcPatches)
{
Info<< "Including all processor patches." << nl << endl;
}
else if (Pstream::parRun())
{
Info<< "Excluding all processor patches." << nl << endl;
}
Info<< "Reading mesh from time " << runTime.value() << endl;
#include "createNamedPolyMesh.H"
// Create local surface from:
// - explicitly named patches only (-patches (at your option)
// - all patches (default in sequential mode)
// - all non-processor patches (default in parallel mode)
// - all non-processor patches (sequential mode, -excludeProcPatches
// (at your option)
// Construct table of patches to include.
const polyBoundaryMesh& bMesh = mesh.boundaryMesh();
labelHashSet includePatches(bMesh.size());
if (args.optionFound("patches"))
{
includePatches = bMesh.patchSet
(
wordReList(args.optionLookup("patches")())
);
}
else
{
forAll(bMesh, patchi)
{
const polyPatch& patch = bMesh[patchi];
if (includeProcPatches || !isA<processorPolyPatch>(patch))
{
includePatches.insert(patchi);
}
}
}
const faceZoneMesh& fzm = mesh.faceZones();
labelHashSet includeFaceZones(fzm.size());
if (args.optionFound("faceZones"))
{
wordReList zoneNames(args.optionLookup("faceZones")());
const wordList allZoneNames(fzm.names());
forAll(zoneNames, i)
{
const wordRe& zoneName = zoneNames[i];
labelList zoneIDs = findStrings(zoneName, allZoneNames);
forAll(zoneIDs, j)
{
includeFaceZones.insert(zoneIDs[j]);
}
if (zoneIDs.empty())
{
WarningInFunction
<< "Cannot find any faceZone name matching "
<< zoneName << endl;
}
}
Info<< "Additionally triangulating faceZones "
<< UIndirectList<word>(allZoneNames, includeFaceZones.sortedToc())
<< endl;
}
// From (name of) patch to compact 'zone' index
HashTable<label> compactZoneID(1000);
// Mesh face and compact zone indx
DynamicList<label> faceLabels;
DynamicList<label> compactZones;
{
// Collect sizes. Hash on names to handle local-only patches (e.g.
// processor patches)
HashTable<label> patchSize(1000);
label nFaces = 0;
forAllConstIter(labelHashSet, includePatches, iter)
{
const polyPatch& pp = bMesh[iter.key()];
patchSize.insert(pp.name(), pp.size());
nFaces += pp.size();
}
HashTable<label> zoneSize(1000);
forAllConstIter(labelHashSet, includeFaceZones, iter)
{
const faceZone& pp = fzm[iter.key()];
zoneSize.insert(pp.name(), pp.size());
nFaces += pp.size();
}
Pstream::mapCombineGather(patchSize, plusEqOp<label>());
Pstream::mapCombineGather(zoneSize, plusEqOp<label>());
// Allocate compact numbering for all patches/faceZones
forAllConstIter(HashTable<label>, patchSize, iter)
{
label sz = compactZoneID.size();
compactZoneID.insert(iter.key(), sz);
}
forAllConstIter(HashTable<label>, zoneSize, iter)
{
label sz = compactZoneID.size();
//Info<< "For faceZone " << iter.key() << " allocating zoneID "
// << sz << endl;
compactZoneID.insert(iter.key(), sz);
}
Pstream::mapCombineScatter(compactZoneID);
// Rework HashTable into labelList just for speed of conversion
labelList patchToCompactZone(bMesh.size(), -1);
labelList faceZoneToCompactZone(bMesh.size(), -1);
forAllConstIter(HashTable<label>, compactZoneID, iter)
{
label patchi = bMesh.findPatchID(iter.key());
if (patchi != -1)
{
patchToCompactZone[patchi] = iter();
}
else
{
label zoneI = fzm.findZoneID(iter.key());
faceZoneToCompactZone[zoneI] = iter();
}
}
faceLabels.setCapacity(nFaces);
compactZones.setCapacity(nFaces);
// Collect faces on patches
forAllConstIter(labelHashSet, includePatches, iter)
{
const polyPatch& pp = bMesh[iter.key()];
forAll(pp, i)
{
faceLabels.append(pp.start()+i);
compactZones.append(patchToCompactZone[pp.index()]);
}
}
// Collect faces on faceZones
forAllConstIter(labelHashSet, includeFaceZones, iter)
{
const faceZone& pp = fzm[iter.key()];
forAll(pp, i)
{
faceLabels.append(pp[i]);
compactZones.append(faceZoneToCompactZone[pp.index()]);
}
}
}
// Addressing engine for all faces
uindirectPrimitivePatch allBoundary
(
UIndirectList<face>(mesh.faces(), faceLabels),
mesh.points()
);
// Find correspondence to master points
labelList pointToGlobal;
labelList uniqueMeshPoints;
autoPtr<globalIndex> globalNumbers = mesh.globalData().mergePoints
(
allBoundary.meshPoints(),
allBoundary.meshPointMap(),
pointToGlobal,
uniqueMeshPoints
);
// Gather all unique points on master
List<pointField> gatheredPoints(Pstream::nProcs());
gatheredPoints[Pstream::myProcNo()] = pointField
(
mesh.points(),
uniqueMeshPoints
);
Pstream::gatherList(gatheredPoints);
// Gather all faces
List<faceList> gatheredFaces(Pstream::nProcs());
gatheredFaces[Pstream::myProcNo()] = allBoundary.localFaces();
forAll(gatheredFaces[Pstream::myProcNo()], i)
{
inplaceRenumber(pointToGlobal, gatheredFaces[Pstream::myProcNo()][i]);
}
Pstream::gatherList(gatheredFaces);
// Gather all ZoneIDs
List<labelList> gatheredZones(Pstream::nProcs());
gatheredZones[Pstream::myProcNo()] = compactZones.xfer();
Pstream::gatherList(gatheredZones);
// On master combine all points, faces, zones
if (Pstream::master())
{
pointField allPoints = ListListOps::combine<pointField>
(
gatheredPoints,
accessOp<pointField>()
);
gatheredPoints.clear();
faceList allFaces = ListListOps::combine<faceList>
(
gatheredFaces,
accessOp<faceList>()
);
gatheredFaces.clear();
labelList allZones = ListListOps::combine<labelList>
(
gatheredZones,
accessOp<labelList>()
);
gatheredZones.clear();
// Zones
surfZoneIdentifierList surfZones(compactZoneID.size());
forAllConstIter(HashTable<label>, compactZoneID, iter)
{
surfZones[iter()] = surfZoneIdentifier(iter.key(), iter());
Info<< "surfZone " << iter() << " : " << surfZones[iter()].name()
<< endl;
}
UnsortedMeshedSurface<face> unsortedFace
(
xferMove(allPoints),
xferMove(allFaces),
xferMove(allZones),
xferMove(surfZones)
);
MeshedSurface<face> sortedFace(unsortedFace);
fileName globalCasePath
(
runTime.processorCase()
? runTime.path()/".."/outFileName
: runTime.path()/outFileName
);
globalCasePath.clean();
Info<< "Writing merged surface to " << globalCasePath << endl;
sortedFace.write(globalCasePath);
}
Info<< "End\n" << endl;
return 0;
}
// ************************************************************************* //