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.
Fixed reaction source terms in the energy and species fraction equations
by multiplying by the phase fraction.
Resolves bug report https://bugs.openfoam.org/view.php?id=2591
Added a grow time and better allocate the CPU time to either add or grow. This
gives much more information to the user and helps changing the settings
accordingly.
Patch contributed by Francesco Contino
vectorField or vector2DField from scalarField components. To do this
properly and have it work for field-type combinations would require some
new field function macros.
Initially the listSwitches functions depended directly on argList functionality
but this has now been factored out so that the listSwitches functions are more
general and require only debug functionality.
Provides better context for the available boundary conditions, fvOptions,
functionObjects etc. and thus returns only those available to and compatible
with the particular application.
e.g.
pimpleFoam -help
Usage: pimpleFoam [OPTIONS]
options:
-case <dir> specify alternate case directory, default is the cwd
-listFunctionObjects
List functionObjects
-listFvOptions List fvOptions
-listRegisteredSwitches
List switches registered for run-time modification
-listScalarBCs List scalar field boundary conditions (fvPatchField<scalar>)
-listSwitches List switches declared in libraries but not set in
etc/controlDict
-listTurbulenceModels
List turbulenceModels
-listUnsetSwitches
List switches declared in libraries but not set in
etc/controlDict
-listVectorBCs List vector field boundary conditions (fvPatchField<vector>)
-noFunctionObjects
do not execute functionObjects
-parallel run in parallel
-postProcess Execute functionObjects only
-roots <(dir1 .. dirN)>
slave root directories for distributed running
-srcDoc display source code in browser
-doc display application documentation in browser
-help print the usage
pimpleFoam listTurbulenceModels
pimpleFoam -listTurbulenceModels
/*---------------------------------------------------------------------------*\
| ========= | |
| \\ / F ield | OpenFOAM: The Open Source CFD Toolbox |
| \\ / O peration | Version: dev |
| \\ / A nd | Web: www.OpenFOAM.org |
| \\/ M anipulation | |
\*---------------------------------------------------------------------------*/
Build : dev-39c46019e44f
Exec : pimpleFoam -listTurbulenceModels
Date : Jun 10 2017
Time : 21:37:49
Host : "dm"
PID : 675
Case : /home/dm2/henry/OpenFOAM/OpenFOAM-dev
nProcs : 1
sigFpe : Enabling floating point exception trapping (FOAM_SIGFPE).
SetNaN : Initialising allocated memory to NaN (FOAM_SETNAN).
fileModificationChecking : Monitoring run-time modified files using timeStampMaster (fileModificationSkew 10)
allowSystemOperations : Allowing user-supplied system call operations
// * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * //
Turbulence models
3
(
LES
RAS
laminar
)
RAS models
18
(
LRR
LamBremhorstKE
LaunderSharmaKE
LienCubicKE
LienLeschziner
RNGkEpsilon
SSG
ShihQuadraticKE
SpalartAllmaras
kEpsilon
kOmega
kOmegaSST
kOmegaSSTLM
kOmegaSSTSAS
kkLOmega
qZeta
realizableKE
v2f
)
LES models
10
(
DeardorffDiffStress
Smagorinsky
SpalartAllmarasDDES
SpalartAllmarasDES
SpalartAllmarasIDDES
WALE
dynamicKEqn
dynamicLagrangian
kEqn
kOmegaSSTDES
)
Further work will be needed to support the -listTurbulenceModels option in
multiphase solvers.
discontinuous fields, with the discontinuity defined by a level set. The
functions do a proper integration of the discontinuous fields by tet-
and tri-cutting along the plane of the level set.
Provides the additional compression necessary to ensure interface integrity
adjacent to a boundary at a low angle of incidence to the interface. This is
particularly important when simulating planing hulls.
Updated the tetrahedron and triangle classes to use the barycentric
primitives. Removed duplicate code for generating random positions in
tets and tris, and fixed bug in tri random position.
This tutorial demonstrates moving mesh and AMI with a Lagrangian cloud.
It is very slow, as interaction lists (required to compute collisions)
are not optimised for moving meshes. The simulation time has therefore
been made very short, so that it finishes in a reasonable time. The
mixer only completes a small fraction of a rotation in this time. This
is still sufficient to test tracking and collisions in the presence of
AMI and mesh motion.
In order to generate a convincing animation, however, the end time must
be increased and the simulation run for a number of days.
and the continuous-phase simulation type
For LTS and steady-state simulations the transient option does not need to be
provided as only steady-state tracking is appropriate. For transient running
the Lagrangian tracking may be steady or transient.
The evolution of a KinematicParcel happens in three stages; (1) tracking
across the cell, (2) interaction with the face or patch that has been
hit, and (3) clculation and and update of parcel and cell properties.
The KinematicParcel used to evolve in this order, as steps 1 and 2 were
part of the same lower level method. This meant that the update stage
was done after interacting with the face, meaning the parcel was not in
the cell that had just been tracked through, or, by means of a patch
interaction, had been modified such that it was no longer representative
of the track through the cell.
With the separation of stages 1 and 2 in the base class, it is now
possible to do the update stage before interacting with the face (i.e.,
proceeding in the order 1, 3, 2). This makes the state consistent for
the updates, and avoids the issues described.
Patch contributed by Timo Niemi, VTT.
This resolves bug report https://bugs.openfoam.org/view.php?id=2282
Particle collisions with ACMI patches are now handled. The hit detects
whether the location is within the overlap or the coupled region and
recurses, calling the hit routine appropriate for the region.
The low level tracking methods are now more consistently named. There is
now a distinction between tracking to a face and hitting it. Function
object side effects have been moved out of the base layer and into the
parcels on which they are meaningful.
The TrackData::switchProcessor flag was not being set for some of the
tracking steps made by the more complicated parcels. In the case that a
parcel starts the step already on a processor boundary, this sometimes
lead to the particle being transferred back and forth indefinitely. The
flag is now explicitly set in all cases.
