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
Contributed by Mattijs Janssens.
1. Any non-blocking data exchange needs to know in advance the sizes to
receive so it can size the buffer. For "halo" exchanges this is not
a problem since the sizes are known in advance but or all other data
exchanges these sizes need to be exchanged in advance.
This was previously done by having all processors send the sizes of data to
send to the master and send it back such that all processors
- had the same information
- all could work out who was sending what to where and hence what needed to
be received.
This is now changed such that we only send the size to the
destination processor (instead of to all as previously). This means
that
- the list of sizes to send is now of size nProcs v.s. nProcs*nProcs before
- we cut out the route to the master and back by using a native MPI
call
It causes a small change to the API of exchange and PstreamBuffers -
they now return the sizes of the local buffers only (a labelList) and
not the sizes of the buffers on all processors (labelListList)
2. Reversing the order of the way in which the sending is done when
scattering information from the master processor to the other
processors. This is done in a tree like fashion. Each processor has a
set of processors to receive from/ send to. When receiving it will
first receive from the processors with the least amount of
sub-processors (i.e. the ones which return first). When sending it
needs to do the opposite: start sending to the processor with the
most amount of sub-tree since this is the critical path.
in case of tmp misuse.
Simplified tmp reuse pattern in field algebra to use tmp copy and
assignment rather than the complex delayed call to 'ptr()'.
Removed support for unused non-const 'REF' storage of non-tmp objects due to C++
limitation in constructor overloading: if both tmp(T&) and tmp(const T&)
constructors are provided resolution is ambiguous.
The turbulence libraries have been upgraded and '-DCONST_TMP' option
specified in the 'options' file to switch to the new 'tmp' behavior.
Now solvers return solver performance information for all components
with backward compatibility provided by the "max" function which created
the scalar solverPerformance from the maximum component residuals from
the SolverPerformance<Type>.
The residuals functionObject has been upgraded to support
SolverPerformance<Type> so that now the initial residuals for all
(valid) components are tabulated, e.g. for the cavity tutorial case the
residuals for p, Ux and Uy are listed vs time.
Currently the residualControl option of pimpleControl and simpleControl
is supported in backward compatibility mode (only the maximum component
residual is considered) but in the future this will be upgraded to
support convergence control for the components individually.
This development started from patches provided by Bruno Santos, See
http://www.openfoam.org/mantisbt/view.php?id=1824
To compile with 64bit labels set
WM_LABEL_SIZE=64
in ~/OpenFOAM/dev/prefs.sh
source ~/.bashrc
then Allwmake in OpenFOAM-dev.
This will build into for example OpenFOAM-dev/platforms/linux64ClangDPInt64Opt
If WM_LABEL_SIZE is unset or set to 32:
WM_LABEL_SIZE=32
the build would be placed into OpenFOAM-dev/platforms/linux64ClangDPInt32Opt
Thus both 32bit and 64bit label builds can coexist without problem.
