OpenFOAM-4.x/src/parallel/decompose/decompositionMethods/simpleGeomDecomp/simpleGeomDecomp.C

466 lines
12 KiB
C

/*---------------------------------------------------------------------------*\
========= |
\\ / F ield | OpenFOAM: The Open Source CFD Toolbox
\\ / O peration |
\\ / A nd | Copyright (C) 2011-2015 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/>.
\*---------------------------------------------------------------------------*/
#include "simpleGeomDecomp.H"
#include "addToRunTimeSelectionTable.H"
#include "SortableList.H"
#include "globalIndex.H"
#include "SubField.H"
// * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * //
namespace Foam
{
defineTypeNameAndDebug(simpleGeomDecomp, 0);
addToRunTimeSelectionTable
(
decompositionMethod,
simpleGeomDecomp,
dictionary
);
}
// * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * //
// assignToProcessorGroup : given nCells cells and nProcGroup processor
// groups to share them, how do we share them out? Answer : each group
// gets nCells/nProcGroup cells, and the first few get one
// extra to make up the numbers. This should produce almost
// perfect load balancing
void Foam::simpleGeomDecomp::assignToProcessorGroup
(
labelList& processorGroup,
const label nProcGroup
) const
{
label jump = processorGroup.size()/nProcGroup;
label jumpb = jump + 1;
label fstProcessorGroup = processorGroup.size() - jump*nProcGroup;
label ind = 0;
label j = 0;
// assign cells to the first few processor groups (those with
// one extra cell each
for (j=0; j<fstProcessorGroup; j++)
{
for (label k=0; k<jumpb; k++)
{
processorGroup[ind++] = j;
}
}
// and now to the `normal' processor groups
for (; j<nProcGroup; j++)
{
for (label k=0; k<jump; k++)
{
processorGroup[ind++] = j;
}
}
}
void Foam::simpleGeomDecomp::assignToProcessorGroup
(
labelList& processorGroup,
const label nProcGroup,
const labelList& indices,
const scalarField& weights,
const scalar summedWeights
) const
{
// This routine gets the sorted points.
// Easiest to explain with an example.
// E.g. 400 points, sum of weights : 513.
// Now with number of divisions in this direction (nProcGroup) : 4
// gives the split at 513/4 = 128
// So summed weight from 0..128 goes into bin 0,
// ,, 128..256 goes into bin 1
// etc.
// Finally any remaining ones go into the last bin (3).
const scalar jump = summedWeights/nProcGroup;
const label nProcGroupM1 = nProcGroup - 1;
scalar sumWeights = 0;
label ind = 0;
label j = 0;
// assign cells to all except last group.
for (j=0; j<nProcGroupM1; j++)
{
const scalar limit = jump*scalar(j + 1);
while (sumWeights < limit)
{
sumWeights += weights[indices[ind]];
processorGroup[ind++] = j;
}
}
// Ensure last included.
while (ind < processorGroup.size())
{
processorGroup[ind++] = nProcGroupM1;
}
}
Foam::labelList Foam::simpleGeomDecomp::decomposeOneProc
(
const pointField& points
) const
{
// construct a list for the final result
labelList finalDecomp(points.size());
labelList processorGroups(points.size());
labelList pointIndices(points.size());
forAll(pointIndices, i)
{
pointIndices[i] = i;
}
const pointField rotatedPoints(rotDelta_ & points);
// and one to take the processor group id's. For each direction.
// we assign the processors to groups of processors labelled
// 0..nX to give a banded structure on the mesh. Then we
// construct the actual processor number by treating this as
// the units part of the processor number.
sort
(
pointIndices,
UList<scalar>::less(rotatedPoints.component(vector::X))
);
assignToProcessorGroup(processorGroups, n_.x());
forAll(points, i)
{
finalDecomp[pointIndices[i]] = processorGroups[i];
}
// now do the same thing in the Y direction. These processor group
// numbers add multiples of nX to the proc. number (columns)
sort
(
pointIndices,
UList<scalar>::less(rotatedPoints.component(vector::Y))
);
assignToProcessorGroup(processorGroups, n_.y());
forAll(points, i)
{
finalDecomp[pointIndices[i]] += n_.x()*processorGroups[i];
}
// finally in the Z direction. Now we add multiples of nX*nY to give
// layers
sort
(
pointIndices,
UList<scalar>::less(rotatedPoints.component(vector::Z))
);
assignToProcessorGroup(processorGroups, n_.z());
forAll(points, i)
{
finalDecomp[pointIndices[i]] += n_.x()*n_.y()*processorGroups[i];
}
return finalDecomp;
}
Foam::labelList Foam::simpleGeomDecomp::decomposeOneProc
(
const pointField& points,
const scalarField& weights
) const
{
// construct a list for the final result
labelList finalDecomp(points.size());
labelList processorGroups(points.size());
labelList pointIndices(points.size());
forAll(pointIndices, i)
{
pointIndices[i] = i;
}
const pointField rotatedPoints(rotDelta_ & points);
// and one to take the processor group id's. For each direction.
// we assign the processors to groups of processors labelled
// 0..nX to give a banded structure on the mesh. Then we
// construct the actual processor number by treating this as
// the units part of the processor number.
sort
(
pointIndices,
UList<scalar>::less(rotatedPoints.component(vector::X))
);
const scalar summedWeights = sum(weights);
assignToProcessorGroup
(
processorGroups,
n_.x(),
pointIndices,
weights,
summedWeights
);
forAll(points, i)
{
finalDecomp[pointIndices[i]] = processorGroups[i];
}
// now do the same thing in the Y direction. These processor group
// numbers add multiples of nX to the proc. number (columns)
sort
(
pointIndices,
UList<scalar>::less(rotatedPoints.component(vector::Y))
);
assignToProcessorGroup
(
processorGroups,
n_.y(),
pointIndices,
weights,
summedWeights
);
forAll(points, i)
{
finalDecomp[pointIndices[i]] += n_.x()*processorGroups[i];
}
// finally in the Z direction. Now we add multiples of nX*nY to give
// layers
sort
(
pointIndices,
UList<scalar>::less(rotatedPoints.component(vector::Z))
);
assignToProcessorGroup
(
processorGroups,
n_.z(),
pointIndices,
weights,
summedWeights
);
forAll(points, i)
{
finalDecomp[pointIndices[i]] += n_.x()*n_.y()*processorGroups[i];
}
return finalDecomp;
}
// * * * * * * * * * * * * * * * * Constructors * * * * * * * * * * * * * * //
Foam::simpleGeomDecomp::simpleGeomDecomp(const dictionary& decompositionDict)
:
geomDecomp(decompositionDict, typeName)
{}
// * * * * * * * * * * * * * * * Member Functions * * * * * * * * * * * * * //
Foam::labelList Foam::simpleGeomDecomp::decompose
(
const pointField& points
)
{
if (!Pstream::parRun())
{
return decomposeOneProc(points);
}
else
{
globalIndex globalNumbers(points.size());
// Collect all points on master
if (Pstream::master())
{
pointField allPoints(globalNumbers.size());
label nTotalPoints = 0;
// Master first
SubField<point>(allPoints, points.size()).assign(points);
nTotalPoints += points.size();
// Add slaves
for (int slave=1; slave<Pstream::nProcs(); slave++)
{
IPstream fromSlave(Pstream::scheduled, slave);
pointField nbrPoints(fromSlave);
SubField<point>
(
allPoints,
nbrPoints.size(),
nTotalPoints
).assign(nbrPoints);
nTotalPoints += nbrPoints.size();
}
// Decompose
labelList finalDecomp(decomposeOneProc(allPoints));
// Send back
for (int slave=1; slave<Pstream::nProcs(); slave++)
{
OPstream toSlave(Pstream::scheduled, slave);
toSlave << SubField<label>
(
finalDecomp,
globalNumbers.localSize(slave),
globalNumbers.offset(slave)
);
}
// Get my own part
finalDecomp.setSize(points.size());
return finalDecomp;
}
else
{
// Send my points
{
OPstream toMaster(Pstream::scheduled, Pstream::masterNo());
toMaster<< points;
}
// Receive back decomposition
IPstream fromMaster(Pstream::scheduled, Pstream::masterNo());
labelList finalDecomp(fromMaster);
return finalDecomp;
}
}
}
Foam::labelList Foam::simpleGeomDecomp::decompose
(
const pointField& points,
const scalarField& weights
)
{
if (!Pstream::parRun())
{
return decomposeOneProc(points, weights);
}
else
{
globalIndex globalNumbers(points.size());
// Collect all points on master
if (Pstream::master())
{
pointField allPoints(globalNumbers.size());
scalarField allWeights(allPoints.size());
label nTotalPoints = 0;
// Master first
SubField<point>(allPoints, points.size()).assign(points);
SubField<scalar>(allWeights, points.size()).assign(weights);
nTotalPoints += points.size();
// Add slaves
for (int slave=1; slave<Pstream::nProcs(); slave++)
{
IPstream fromSlave(Pstream::scheduled, slave);
pointField nbrPoints(fromSlave);
scalarField nbrWeights(fromSlave);
SubField<point>
(
allPoints,
nbrPoints.size(),
nTotalPoints
).assign(nbrPoints);
SubField<scalar>
(
allWeights,
nbrWeights.size(),
nTotalPoints
).assign(nbrWeights);
nTotalPoints += nbrPoints.size();
}
// Decompose
labelList finalDecomp(decomposeOneProc(allPoints, allWeights));
// Send back
for (int slave=1; slave<Pstream::nProcs(); slave++)
{
OPstream toSlave(Pstream::scheduled, slave);
toSlave << SubField<label>
(
finalDecomp,
globalNumbers.localSize(slave),
globalNumbers.offset(slave)
);
}
// Get my own part
finalDecomp.setSize(points.size());
return finalDecomp;
}
else
{
// Send my points
{
OPstream toMaster(Pstream::scheduled, Pstream::masterNo());
toMaster<< points << weights;
}
// Receive back decomposition
IPstream fromMaster(Pstream::scheduled, Pstream::masterNo());
labelList finalDecomp(fromMaster);
return finalDecomp;
}
}
}
// ************************************************************************* //