7c301dbff4
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.
745 lines
17 KiB
C
745 lines
17 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/>.
|
|
|
|
\*---------------------------------------------------------------------------*/
|
|
|
|
#include "functionObjectList.H"
|
|
#include "Time.H"
|
|
#include "mapPolyMesh.H"
|
|
#include "argList.H"
|
|
#include "timeControlFunctionObject.H"
|
|
//#include "IFstream.H"
|
|
#include "dictionaryEntry.H"
|
|
#include "stringOps.H"
|
|
#include "Tuple2.H"
|
|
#include "etcFiles.H"
|
|
#include "IOdictionary.H"
|
|
|
|
/* * * * * * * * * * * * * * * Static Member Data * * * * * * * * * * * * * */
|
|
|
|
Foam::fileName Foam::functionObjectList::functionObjectDictPath
|
|
(
|
|
"caseDicts/postProcessing"
|
|
);
|
|
|
|
|
|
// * * * * * * * * * * * * Private Member Functions * * * * * * * * * * * * //
|
|
|
|
Foam::functionObject* Foam::functionObjectList::remove
|
|
(
|
|
const word& key,
|
|
label& oldIndex
|
|
)
|
|
{
|
|
functionObject* ptr = 0;
|
|
|
|
// Find index of existing functionObject
|
|
HashTable<label>::iterator fnd = indices_.find(key);
|
|
|
|
if (fnd != indices_.end())
|
|
{
|
|
oldIndex = fnd();
|
|
|
|
// Retrieve the pointer and remove it from the old list
|
|
ptr = this->set(oldIndex, 0).ptr();
|
|
indices_.erase(fnd);
|
|
}
|
|
else
|
|
{
|
|
oldIndex = -1;
|
|
}
|
|
|
|
return ptr;
|
|
}
|
|
|
|
|
|
void Foam::functionObjectList::listDir
|
|
(
|
|
const fileName& dir,
|
|
HashSet<word>& foMap
|
|
)
|
|
{
|
|
// Search specified directory for functionObject configuration files
|
|
{
|
|
fileNameList foFiles(fileHandler().readDir(dir));
|
|
forAll(foFiles, f)
|
|
{
|
|
if (foFiles[f].ext().empty())
|
|
{
|
|
foMap.insert(foFiles[f]);
|
|
}
|
|
}
|
|
}
|
|
|
|
// Recurse into sub-directories
|
|
{
|
|
fileNameList foDirs(fileHandler().readDir(dir, fileName::DIRECTORY));
|
|
forAll(foDirs, fd)
|
|
{
|
|
listDir(dir/foDirs[fd], foMap);
|
|
}
|
|
}
|
|
}
|
|
|
|
|
|
void Foam::functionObjectList::list()
|
|
{
|
|
HashSet<word> foMap;
|
|
|
|
fileNameList etcDirs(findEtcDirs(functionObjectDictPath));
|
|
|
|
forAll(etcDirs, ed)
|
|
{
|
|
listDir(etcDirs[ed], foMap);
|
|
}
|
|
|
|
Info<< nl
|
|
<< "Available configured functionObjects:"
|
|
<< foMap.sortedToc()
|
|
<< nl;
|
|
}
|
|
|
|
|
|
Foam::fileName Foam::functionObjectList::findDict(const word& funcName)
|
|
{
|
|
// First check if there is a functionObject dictionary file in the
|
|
// case system directory
|
|
fileName dictFile = stringOps::expand("$FOAM_CASE")/"system"/funcName;
|
|
|
|
if (isFile(dictFile))
|
|
{
|
|
return dictFile;
|
|
}
|
|
else
|
|
{
|
|
fileNameList etcDirs(findEtcDirs(functionObjectDictPath));
|
|
|
|
forAll(etcDirs, i)
|
|
{
|
|
dictFile = search(funcName, etcDirs[i]);
|
|
if (!dictFile.empty())
|
|
{
|
|
return dictFile;
|
|
}
|
|
}
|
|
}
|
|
|
|
return fileName::null;
|
|
}
|
|
|
|
|
|
bool Foam::functionObjectList::readFunctionObject
|
|
(
|
|
const string& funcNameArgs,
|
|
dictionary& functionsDict,
|
|
HashSet<word>& requiredFields,
|
|
const word& region
|
|
)
|
|
{
|
|
// Parse the optional functionObject arguments:
|
|
// 'Q(U)' -> funcName = Q; args = (U); field = U
|
|
//
|
|
// Supports named arguments:
|
|
// 'patchAverage(patch=inlet, p)' -> funcName = patchAverage;
|
|
// args = (patch=inlet, p); field = p
|
|
|
|
word funcName(funcNameArgs);
|
|
|
|
int argLevel = 0;
|
|
wordList args;
|
|
|
|
List<Tuple2<word, string>> namedArgs;
|
|
bool namedArg = false;
|
|
word argName;
|
|
|
|
word::size_type start = 0;
|
|
word::size_type i = 0;
|
|
|
|
for
|
|
(
|
|
word::const_iterator iter = funcNameArgs.begin();
|
|
iter != funcNameArgs.end();
|
|
++iter
|
|
)
|
|
{
|
|
char c = *iter;
|
|
|
|
if (c == '(')
|
|
{
|
|
if (argLevel == 0)
|
|
{
|
|
funcName = funcNameArgs(start, i - start);
|
|
start = i+1;
|
|
}
|
|
++argLevel;
|
|
}
|
|
else if (c == ',' || c == ')')
|
|
{
|
|
if (argLevel == 1)
|
|
{
|
|
if (namedArg)
|
|
{
|
|
namedArgs.append
|
|
(
|
|
Tuple2<word, string>
|
|
(
|
|
argName,
|
|
funcNameArgs(start, i - start)
|
|
)
|
|
);
|
|
namedArg = false;
|
|
}
|
|
else
|
|
{
|
|
args.append
|
|
(
|
|
string::validate<word>(funcNameArgs(start, i - start))
|
|
);
|
|
}
|
|
start = i+1;
|
|
}
|
|
|
|
if (c == ')')
|
|
{
|
|
if (argLevel == 1)
|
|
{
|
|
break;
|
|
}
|
|
--argLevel;
|
|
}
|
|
}
|
|
else if (c == '=')
|
|
{
|
|
argName = string::validate<word>(funcNameArgs(start, i - start));
|
|
start = i+1;
|
|
namedArg = true;
|
|
}
|
|
|
|
++i;
|
|
}
|
|
|
|
// Search for the functionObject dictionary
|
|
fileName path = findDict(funcName);
|
|
|
|
if (path == fileName::null)
|
|
{
|
|
WarningInFunction
|
|
<< "Cannot find functionObject file " << funcName << endl;
|
|
return false;
|
|
}
|
|
|
|
// Read the functionObject dictionary
|
|
//IFstream fileStream(path);
|
|
autoPtr<ISstream> fileStreamPtr(fileHandler().NewIFstream(path));
|
|
ISstream& fileStream = fileStreamPtr();
|
|
|
|
dictionary funcsDict(fileStream);
|
|
dictionary* funcDictPtr = &funcsDict;
|
|
|
|
if (funcsDict.found(funcName) && funcsDict.isDict(funcName))
|
|
{
|
|
funcDictPtr = &funcsDict.subDict(funcName);
|
|
}
|
|
|
|
dictionary& funcDict = *funcDictPtr;
|
|
|
|
// Insert the 'field' and/or 'fields' entry corresponding to the optional
|
|
// arguments or read the 'field' or 'fields' entry and add the required
|
|
// fields to requiredFields
|
|
if (args.size() == 1)
|
|
{
|
|
funcDict.set("field", args[0]);
|
|
funcDict.set("fields", args);
|
|
requiredFields.insert(args[0]);
|
|
}
|
|
else if (args.size() > 1)
|
|
{
|
|
funcDict.set("fields", args);
|
|
requiredFields.insert(args);
|
|
}
|
|
else if (funcDict.found("field"))
|
|
{
|
|
requiredFields.insert(word(funcDict.lookup("field")));
|
|
}
|
|
else if (funcDict.found("fields"))
|
|
{
|
|
requiredFields.insert(wordList(funcDict.lookup("fields")));
|
|
}
|
|
|
|
// Insert named arguments
|
|
forAll(namedArgs, i)
|
|
{
|
|
IStringStream entryStream
|
|
(
|
|
namedArgs[i].first() + ' ' + namedArgs[i].second() + ';'
|
|
);
|
|
funcDict.set(entry::New(entryStream).ptr());
|
|
}
|
|
|
|
// Insert the region name if specified
|
|
if (region != word::null)
|
|
{
|
|
funcDict.set("region", region);
|
|
}
|
|
|
|
// Merge this functionObject dictionary into functionsDict
|
|
dictionary funcArgsDict;
|
|
funcArgsDict.add(string::validate<word>(funcNameArgs), funcDict);
|
|
functionsDict.merge(funcArgsDict);
|
|
|
|
return true;
|
|
}
|
|
|
|
|
|
// * * * * * * * * * * * * * * * * Constructors * * * * * * * * * * * * * * //
|
|
|
|
Foam::functionObjectList::functionObjectList
|
|
(
|
|
const Time& t,
|
|
const bool execution
|
|
)
|
|
:
|
|
PtrList<functionObject>(),
|
|
digests_(),
|
|
indices_(),
|
|
time_(t),
|
|
parentDict_(t.controlDict()),
|
|
execution_(execution),
|
|
updated_(false)
|
|
{}
|
|
|
|
|
|
Foam::functionObjectList::functionObjectList
|
|
(
|
|
const Time& t,
|
|
const dictionary& parentDict,
|
|
const bool execution
|
|
)
|
|
:
|
|
PtrList<functionObject>(),
|
|
digests_(),
|
|
indices_(),
|
|
time_(t),
|
|
parentDict_(parentDict),
|
|
execution_(execution),
|
|
updated_(false)
|
|
{}
|
|
|
|
|
|
Foam::autoPtr<Foam::functionObjectList> Foam::functionObjectList::New
|
|
(
|
|
const argList& args,
|
|
const Time& runTime,
|
|
dictionary& controlDict,
|
|
HashSet<word>& requiredFields
|
|
)
|
|
{
|
|
autoPtr<functionObjectList> functionsPtr;
|
|
|
|
controlDict.add
|
|
(
|
|
dictionaryEntry("functions", controlDict, dictionary::null)
|
|
);
|
|
|
|
dictionary& functionsDict = controlDict.subDict("functions");
|
|
|
|
word region = word::null;
|
|
|
|
// Set the region name if specified
|
|
if (args.optionFound("region"))
|
|
{
|
|
region = args["region"];
|
|
}
|
|
|
|
if
|
|
(
|
|
args.optionFound("dict")
|
|
|| args.optionFound("func")
|
|
|| args.optionFound("funcs")
|
|
)
|
|
{
|
|
if (args.optionFound("dict"))
|
|
{
|
|
controlDict.merge
|
|
(
|
|
IOdictionary
|
|
(
|
|
IOobject
|
|
(
|
|
args["dict"],
|
|
runTime,
|
|
IOobject::MUST_READ_IF_MODIFIED
|
|
)
|
|
)
|
|
);
|
|
}
|
|
|
|
if (args.optionFound("func"))
|
|
{
|
|
readFunctionObject
|
|
(
|
|
args["func"],
|
|
functionsDict,
|
|
requiredFields,
|
|
region
|
|
);
|
|
}
|
|
|
|
if (args.optionFound("funcs"))
|
|
{
|
|
wordList funcs(args.optionLookup("funcs")());
|
|
|
|
forAll(funcs, i)
|
|
{
|
|
readFunctionObject
|
|
(
|
|
funcs[i],
|
|
functionsDict,
|
|
requiredFields,
|
|
region
|
|
);
|
|
}
|
|
}
|
|
|
|
functionsPtr.reset(new functionObjectList(runTime, controlDict));
|
|
}
|
|
else
|
|
{
|
|
functionsPtr.reset(new functionObjectList(runTime));
|
|
}
|
|
|
|
functionsPtr->start();
|
|
|
|
return functionsPtr;
|
|
}
|
|
|
|
|
|
// * * * * * * * * * * * * * * * * Destructor * * * * * * * * * * * * * * * //
|
|
|
|
Foam::functionObjectList::~functionObjectList()
|
|
{}
|
|
|
|
|
|
// * * * * * * * * * * * * * * * Member Functions * * * * * * * * * * * * * //
|
|
|
|
void Foam::functionObjectList::clear()
|
|
{
|
|
PtrList<functionObject>::clear();
|
|
digests_.clear();
|
|
indices_.clear();
|
|
updated_ = false;
|
|
}
|
|
|
|
|
|
Foam::label Foam::functionObjectList::findObjectID(const word& name) const
|
|
{
|
|
forAll(*this, objectI)
|
|
{
|
|
if (operator[](objectI).name() == name)
|
|
{
|
|
return objectI;
|
|
}
|
|
}
|
|
|
|
return -1;
|
|
}
|
|
|
|
|
|
void Foam::functionObjectList::on()
|
|
{
|
|
execution_ = true;
|
|
}
|
|
|
|
|
|
void Foam::functionObjectList::off()
|
|
{
|
|
// For safety, also force a read() when execution is turned back on
|
|
updated_ = execution_ = false;
|
|
}
|
|
|
|
|
|
bool Foam::functionObjectList::status() const
|
|
{
|
|
return execution_;
|
|
}
|
|
|
|
|
|
bool Foam::functionObjectList::start()
|
|
{
|
|
return read();
|
|
}
|
|
|
|
|
|
bool Foam::functionObjectList::execute()
|
|
{
|
|
bool ok = true;
|
|
|
|
if (execution_)
|
|
{
|
|
if (!updated_)
|
|
{
|
|
read();
|
|
}
|
|
|
|
forAll(*this, objectI)
|
|
{
|
|
ok = operator[](objectI).execute() && ok;
|
|
ok = operator[](objectI).write() && ok;
|
|
}
|
|
}
|
|
|
|
return ok;
|
|
}
|
|
|
|
|
|
bool Foam::functionObjectList::end()
|
|
{
|
|
bool ok = true;
|
|
|
|
if (execution_)
|
|
{
|
|
if (!updated_)
|
|
{
|
|
read();
|
|
}
|
|
|
|
forAll(*this, objectI)
|
|
{
|
|
ok = operator[](objectI).end() && ok;
|
|
}
|
|
}
|
|
|
|
return ok;
|
|
}
|
|
|
|
|
|
bool Foam::functionObjectList::adjustTimeStep()
|
|
{
|
|
bool ok = true;
|
|
|
|
if (execution_)
|
|
{
|
|
if (!updated_)
|
|
{
|
|
read();
|
|
}
|
|
|
|
forAll(*this, objectI)
|
|
{
|
|
ok = operator[](objectI).adjustTimeStep() && ok;
|
|
}
|
|
}
|
|
|
|
return ok;
|
|
}
|
|
|
|
|
|
bool Foam::functionObjectList::read()
|
|
{
|
|
bool ok = true;
|
|
updated_ = execution_;
|
|
|
|
// Avoid reading/initializing if execution is off
|
|
if (!execution_)
|
|
{
|
|
return true;
|
|
}
|
|
|
|
// Update existing and add new functionObjects
|
|
const entry* entryPtr = parentDict_.lookupEntryPtr
|
|
(
|
|
"functions",
|
|
false,
|
|
false
|
|
);
|
|
|
|
if (entryPtr)
|
|
{
|
|
PtrList<functionObject> newPtrs;
|
|
List<SHA1Digest> newDigs;
|
|
HashTable<label> newIndices;
|
|
|
|
label nFunc = 0;
|
|
|
|
if (!entryPtr->isDict())
|
|
{
|
|
FatalIOErrorInFunction(parentDict_)
|
|
<< "'functions' entry is not a dictionary"
|
|
<< exit(FatalIOError);
|
|
}
|
|
|
|
const dictionary& functionsDict = entryPtr->dict();
|
|
|
|
const_cast<Time&>(time_).libs().open
|
|
(
|
|
functionsDict,
|
|
"libs",
|
|
functionObject::dictionaryConstructorTablePtr_
|
|
);
|
|
|
|
newPtrs.setSize(functionsDict.size());
|
|
newDigs.setSize(functionsDict.size());
|
|
|
|
forAllConstIter(dictionary, functionsDict, iter)
|
|
{
|
|
const word& key = iter().keyword();
|
|
|
|
if (!iter().isDict())
|
|
{
|
|
if (key != "libs")
|
|
{
|
|
IOWarningInFunction(parentDict_)
|
|
<< "Entry " << key << " is not a dictionary" << endl;
|
|
}
|
|
|
|
continue;
|
|
}
|
|
|
|
const dictionary& dict = iter().dict();
|
|
bool enabled = dict.lookupOrDefault("enabled", true);
|
|
|
|
newDigs[nFunc] = dict.digest();
|
|
|
|
label oldIndex;
|
|
functionObject* objPtr = remove(key, oldIndex);
|
|
|
|
if (objPtr)
|
|
{
|
|
if (enabled)
|
|
{
|
|
// Dictionary changed for an existing functionObject
|
|
if (newDigs[nFunc] != digests_[oldIndex])
|
|
{
|
|
ok = objPtr->read(dict) && ok;
|
|
}
|
|
}
|
|
else
|
|
{
|
|
// Delete the disabled functionObject
|
|
delete objPtr;
|
|
objPtr = nullptr;
|
|
continue;
|
|
}
|
|
}
|
|
else if (enabled)
|
|
{
|
|
autoPtr<functionObject> foPtr;
|
|
|
|
FatalError.throwExceptions();
|
|
FatalIOError.throwExceptions();
|
|
try
|
|
{
|
|
if
|
|
(
|
|
dict.found("writeControl")
|
|
|| dict.found("outputControl")
|
|
)
|
|
{
|
|
foPtr.set
|
|
(
|
|
new functionObjects::timeControl(key, time_, dict)
|
|
);
|
|
}
|
|
else
|
|
{
|
|
foPtr = functionObject::New(key, time_, dict);
|
|
}
|
|
}
|
|
catch (Foam::IOerror& ioErr)
|
|
{
|
|
Info<< ioErr << nl << endl;
|
|
::exit(1);
|
|
}
|
|
catch (Foam::error& err)
|
|
{
|
|
WarningInFunction
|
|
<< "Caught FatalError " << err << nl << endl;
|
|
}
|
|
FatalError.dontThrowExceptions();
|
|
FatalIOError.dontThrowExceptions();
|
|
|
|
if (foPtr.valid())
|
|
{
|
|
objPtr = foPtr.ptr();
|
|
}
|
|
else
|
|
{
|
|
ok = false;
|
|
}
|
|
}
|
|
|
|
// Insert active functionObjects into the list
|
|
if (objPtr)
|
|
{
|
|
newPtrs.set(nFunc, objPtr);
|
|
newIndices.insert(key, nFunc);
|
|
nFunc++;
|
|
}
|
|
}
|
|
|
|
newPtrs.setSize(nFunc);
|
|
newDigs.setSize(nFunc);
|
|
|
|
// Updating the PtrList of functionObjects deletes any
|
|
// existing unused functionObjects
|
|
PtrList<functionObject>::transfer(newPtrs);
|
|
digests_.transfer(newDigs);
|
|
indices_.transfer(newIndices);
|
|
}
|
|
else
|
|
{
|
|
PtrList<functionObject>::clear();
|
|
digests_.clear();
|
|
indices_.clear();
|
|
}
|
|
|
|
return ok;
|
|
}
|
|
|
|
|
|
void Foam::functionObjectList::updateMesh(const mapPolyMesh& mpm)
|
|
{
|
|
if (execution_)
|
|
{
|
|
forAll(*this, objectI)
|
|
{
|
|
operator[](objectI).updateMesh(mpm);
|
|
}
|
|
}
|
|
}
|
|
|
|
|
|
void Foam::functionObjectList::movePoints(const polyMesh& mesh)
|
|
{
|
|
if (execution_)
|
|
{
|
|
forAll(*this, objectI)
|
|
{
|
|
operator[](objectI).movePoints(mesh);
|
|
}
|
|
}
|
|
}
|
|
|
|
|
|
// ************************************************************************* //
|