Now faceZones are handled directly by the applications and the new
faceZone::topoChange function so that any face can now be in any number of
zones, significantly increasing the flexibility and usefulness of faceZones.
This completes the generalisation of cellZone, faceZone and pointZone to support
multiple zones for each cell, face or point respectively. Next step will be to
make zones polymorphic and run-time selectable so that they can alter during the
run and adapt to moving meshes for example.
Zones are now completely dynamic, i.e. the number of zones of each type can
change during the run, e.g. by run-time mesh-to-mesh mapping onto meshes with
different zones used to control mesh motion. This means that the index of each
zone may change during the run and so it better that the zones do not cache
their own index but it is looked-up from the zone list using findIndex when
required.
This avoids potential hidden run-time errors caused by solvers running with
boundary conditions which are not fully specified. Note that "null-constructor"
here means the constructor from patch and internal field only, no data is
provided.
Constraint and simple BCs such as 'calculated', 'zeroGradient' and others which
do not require user input to fully specify their operation remain on the
null-constructor table for the construction of fields with for example all
'calculated' or all 'zeroGradient' BCs.
A special version of the 'inletOutlet' fvPatchField named 'zeroInletOutlet' has
been added in which the inlet value is hard-coded to zero which allows this BC
to be included on the null-constructor table. This is useful for the 'age'
functionObject to avoid the need to provide the 'age' volScalarField at time 0
unless special inlet or outlet BCs are required. Also for isothermalFilm in
which the 'alpha' field is created automatically from the 'delta' field if it is
not present and can inherit 'zeroInletOutlet' from 'delta' if appropriate. If a
specific 'inletValue' is require or other more complex BCs then the 'alpha'
field file must be provided to specify these BCs as before.
Following this improvement it will now be possible to remove the
null-constructors from all fvPatchFields not added to the null-constructor
table, which is most of them, thus reducing the amount of code and maintenance
overhead and making easier and more obvious to write new fvPatchField types.
The timeName() function simply returns the dimensionedScalar::name() which holds
the user-time name of the current time and now that timeName() is no longer
virtual the dimensionedScalar::name() can be called directly. The timeName()
function implementation is maintained for backward-compatibility.
The mappedPatchBase has been separated into a type which maps from
another patch (still called mappedPatchBase) and one that maps from
internal cell values (mappedInternalPatchBase). This prevents the user
needing to specify settings for mapping procedures that are not being
used, and potentially don't even make sense given the context in which
they are being applied. It also removes a lot of fragile logic and error
states in the mapping engine and its derivatives regarding the mode of
operation. Mapping from any face in the boundary is no longer supported.
Most region-coupling mapping patches are generated automatically by
utilities like splitMeshRegions and extrudeToRegionMesh. Cases which
create region-coupling mapped patches in this way will likely require no
modification.
Explicitly user-specified mapping will need modifying, however. For
example, where an inlet boundary is mapped to a downstream position in
order to evolve a developed profile. Or if a multi-region simulation is
constructed manually, without using one of the region-generating
utilities.
The available mapped patch types are now as follows:
- mapped: Maps values from one patch to another. Typically used for
inlets and outlets; to map values from an outlet patch to an inlet
patch in order to evolve a developed inlet profile, or to permit
flow between regions. Example specification in blockMesh:
inlet
{
type mapped;
neighbourRegion region0; // Optional. Defaults to the same
// region as the patch.
neighbourPatch outlet;
faces ( ... );
}
Note that any transformation between the patches is now determined
automatically. Alternatively, it can be explicitly specified using
the same syntax as for cyclic patches. The "offset" and "distance"
keywords are no longer used.
- mappedWall: As mapped, but treated as a wall for the purposes of
modelling (wall distance). No transformation. Typically used for
thermally coupling different regions. Usually created automatically
by meshing utilities. Example:
fluid_to_solid
{
type mappedWall;
neighbourRegion solid;
neighbourPatch solid_to_fluid;
method intersection; // The patchToPatch method. See
// below.
faces ( ... );
}
- mappedExtrudedWall: As mapped wall, but with corrections to account
for the thickness of an extruded mesh. Used for region coupling
involving film and thermal baffle models. Almost always generated
automatically by extrudeToRegionMesh (so no example given).
- mappedInternal: Map values from internal cells to a patch. Typically
used for inlets; to map values from internal cells to the inlet in
order to evolve a developed inlet profile. Example:
inlet
{
type mappedInternal;
distance 0.05; // Normal distance from the patch
// from which to map cell values
//offset (0.05 0 0); // Offset from the patch from
// which to map cell values
faces ( ... );
}
Note that an "offsetMode" entry is no longer necessary. The mode
will be inferred from the presence of the distance or offset
entries. If both are provided, then offsetMode will also be required
to choose which setting applies.
The mapped, mappedWall and mappedExtrudedWall patches now permit
specification of a "method". This selects a patchToPatch object and
therefore determines how values are transferred or interpolated between
the patches. Valid options are:
- nearest: Copy the value from the nearest face in the neighbouring
patch.
- matching: As nearest, but with checking to make sure that the
mapping is one-to-one. This is appropriate for patches that are
identically meshed.
- inverseDistance: Inverse distance weighting from a small stencil of
nearby faces in the neighbouring patch.
- intersection: Weighting based on the overlapping areas with faces in
the neighbouring patch. Equivalent to the previous AMI-based mapping
mode.
If a method is not specfied, then the pre-existing approach will apply.
This should be equivalent to the "nearest" method (though in most such
cases, "matching" is probably more appropriate). This fallback may be
removed in the future once the patchToPatch methods have been proven
robust.
The important mapped boundary conditions are now as follows:
- mappedValue: Maps values from one patch to another, and optionally
modify the mapped values to recover a specified average. Example:
inlet
{
type mappedValue;
field U; // Optional. Defaults to the same
// as this field.
average (10 0 0); // The presence of this entry now
// enables setting of the average,
// so "setAverage" is not needed
value uniform 0.1;
}
- mappedInternalValue: Map values from cells to a patch, and
optionally specify the average as in mappedValue. Example:
inlet
{
type mappedValue;
field k; // Optional. Defaults to the same
// as this field.
interpolationScheme cell;
value uniform 0.1;
}
- mappedFlowRateVelocity: Maps the flow rate from one patch to
another, and use this to set a patch-normal velocity. Example:
inlet
{
type mappedFlowRate;
value uniform (0 0 0);
}
Of these, mappedValue and mappedInternalValue can override the
underlying mapped patch's settings by additionally specifying mapping
information (i.e., the neighbourPatch, offset, etc... settings usually
supplied for the patch). This also means these boundary condtions can be
applied to non-mapped patches. This functionality used to be provided
with a separate "mappedField" boundary condition, which has been removed
as it is no longer necessary.
Other mapped boundary conditions are either extremely niche (e.g.,
mappedVelocityFlux), are always automatically generated (e.g.,
mappedValueAndPatchInternalValue), or their usage has not changed (e.g.,
compressible::turbulentTemperatureCoupledBaffleMixed and
compressible::turbulentTemperatureRadCoupledMixed). Use foamInfo to
obtain further details about these conditions.
Poly patches should not hold non-uniform physical data that needs
mapping on mesh changes (decomposition, reconstruction, topology change,
etc ...). They should only hold uniform data that can be user-specified,
or non-uniform data that can be constructed on the fly from the poly
mesh.
With the recent changes to mappedPatchBase and extrudeToRegionMesh, this
has now been consistenly enforced, and a number of incomplete
implementations of poly patch mapping have therefore been removed.
This new mapping structure is designed to support run-time mesh-to-mesh mapping
to allow arbitrary changes to the mesh structure, for example during extreme
motion requiring significant topology change including region disconnection etc.
The polyTopoChangeMap is the map specifically relating to polyMesh topological
changes generated by polyTopoChange and used to update and map mesh related
types and fields following the topo-change.
boundaryProcAddressing has been removed. This has not been needed for a
long time. decomposePar has been optimised for mininum IO, rather than
minimum memory usage. decomposePar has also been corrected so that it
can decompose sequences of time-varying meshes.
used to check the existence of and open an object file, read and check the
header without constructing the object.
'typeIOobject' operates in an equivalent and consistent manner to 'regIOobject'
but the type information is provided by the template argument rather than via
virtual functions for which the derived object would need to be constructed,
which is the case for 'regIOobject'.
'typeIOobject' replaces the previous separate functions 'typeHeaderOk' and
'typeFilePath' with a single consistent interface.
now all path functions in 'IOobject' are either templated on the type or require a
'globalFile' argument to specify if the type is case global e.g. 'IOdictionary' or
decomposed in parallel, e.g. almost everything else.
The 'global()' and 'globalFile()' virtual functions are now in 'regIOobject'
abstract base-class and overridden as required by derived classes. The path
functions using 'global()' and 'globalFile()' to differentiate between global
and processor local objects are now also in 'regIOobject' rather than 'IOobject'
to ensure the path returned is absolutely consistent with the type.
Unfortunately there is still potential for unexpected IO behaviour inconsistent
with the global/local nature of the type due to the 'fileOperation' classes
searching the processor directory for case global objects before searching the
case directory. This approach appears to be a work-around for incomplete
integration with and rationalisation of 'IOobject' but with the changes above it
is no longer necessary. Unfortunately this "up" searching is baked-in at a low
level and mixed-up with various complex ways to pick the processor directory
name out of the object path and will take some unravelling but this work will
undertaken as time allows.
For many information and diagnostic messages the absolute path of the object is
not required and the local path relative to the current case is sufficient; the
new localObjectPath() member function of IOobject provides a convenient way of
printing this.
Registration occurs when the temporary field is transferred to a non-temporary
field via a constructor or if explicitly transferred to the database via the
regIOobject "store" methods.
Changed the default region name from "domain" to "region" for consistency with
the rest of OpenFOAM.
Changed the multiple default region numbering to start from 1 rather than 0
because the top-level mesh in the case directory is always named "region0".
Changed the default region numbering to only relate to the default named regions
and does not increment for explicitly named regions. This avoids a naming
dependency on the default and named region order.
Added new option "-defaultRegionName <name>"
to specify the base name of the unspecified regions, defaults to "region"
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.
splitMeshRegions: handle flipping of faces for surface fields
subsetMesh: subset dimensionedFields
decomposePar: use run-time selection of decomposition constraints. Used to
keep cells on particular processors. See the decomposeParDict in
$FOAM_UTILITIES/parallel/decomposePar:
- preserveBaffles: keep baffle faces on same processor
- preserveFaceZones: keep faceZones owner and neighbour on same processor
- preservePatches: keep owner and neighbour on same processor. Note: not
suitable for cyclicAMI since these are not coupled on the patch level
- singleProcessorFaceSets: keep complete faceSet on a single processor
- refinementHistory: keep cells originating from a single cell on the
same processor.
decomposePar: clean up decomposition of refinement data from snappyHexMesh
reconstructPar: reconstruct refinement data (refineHexMesh, snappyHexMesh)
reconstructParMesh: reconstruct refinement data (refineHexMesh, snappyHexMesh)
redistributePar:
- corrected mapping surfaceFields
- adding processor patches in order consistent with decomposePar
argList: check that slaves are running same version as master
fvMeshSubset: move to dynamicMesh library
fvMeshDistribute:
- support for mapping dimensionedFields
- corrected mapping of surfaceFields
parallel routines: allow parallel running on single processor
Field: support for
- distributed mapping
- mapping with flipping
mapDistribute: support for flipping
AMIInterpolation: avoid constructing localPoints
The deprecated non-const tmp functionality is now on the compiler switch
NON_CONST_TMP which can be enabled by adding -DNON_CONST_TMP to EXE_INC
in the Make/options file. However, it is recommended to upgrade all
code to the new safer tmp by using the '.ref()' member function rather
than the non-const '()' dereference operator when non-const access to
the temporary object is required.
Please report any problems on Mantis.
Henry G. Weller
CFD Direct.
