1 // Copyright (C) 2007-2015 CEA/DEN, EDF R&D
3 // This library is free software; you can redistribute it and/or
4 // modify it under the terms of the GNU Lesser General Public
5 // License as published by the Free Software Foundation; either
6 // version 2.1 of the License, or (at your option) any later version.
8 // This library is distributed in the hope that it will be useful,
9 // but WITHOUT ANY WARRANTY; without even the implied warranty of
10 // MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the GNU
11 // Lesser General Public License for more details.
13 // You should have received a copy of the GNU Lesser General Public
14 // License along with this library; if not, write to the Free Software
15 // Foundation, Inc., 59 Temple Place, Suite 330, Boston, MA 02111-1307 USA
17 // See http://www.salome-platform.org/ or email : webmaster.salome@opencascade.com
21 #include "CommInterface.hxx"
22 #include "Topology.hxx"
23 #include "BlockTopology.hxx"
24 #include "ComponentTopology.hxx"
25 #include "ParaFIELD.hxx"
26 #include "MPIProcessorGroup.hxx"
27 #include "ParaMESH.hxx"
29 #include "InterpolationMatrix.hxx"
30 #include "InterpKernelDEC.hxx"
31 #include "ElementLocator.hxx"
37 \anchor InterpKernelDEC-det
38 \class InterpKernelDEC
40 \section dec-over Overview
42 The InterpKernelDEC enables the \ref InterpKerRemapGlobal "remapping" of fields between two parallel codes.
43 This remapping is based on the computation of intersection volumes between elements from code A
44 and elements from code B. The computation is possible for 3D meshes, 2D meshes, and 3D-surface
45 meshes. Dimensions must be similar for code A and code B (for instance, though it could be
46 desirable, it is not yet possible to couple 3D surfaces with 2D surfaces).
48 In the present version, only fields lying on elements are considered.
50 \image html NonCoincident_small.png "Example showing the transfer from a field based on a
51 quadrangular mesh to a triangular mesh. In a P0-P0 interpolation, to obtain the value on a triangle,
52 the values on quadrangles are weighted by their intersection area and summed."
54 \image latex NonCoincident_small.eps "Example showing the transfer from a field based on a quadrangular
55 mesh to a triangular mesh. In a P0-P0 interpolation, to obtain the value on a triangle, the values
56 on quadrangles are weighted by their intersection area and summed."
58 A typical use of InterpKernelDEC encompasses two distinct phases :
59 - A setup phase during which the intersection volumes are computed and the communication structures are
60 setup. This corresponds to calling the InterpKernelDEC::synchronize() method.
61 - A use phase during which the remappings are actually performed. This corresponds to the calls to
62 sendData() and recvData() which actually trigger the data exchange. The data exchange are synchronous
63 in the current version of the library so that recvData() and sendData() calls must be synchronized
64 on code A and code B processor groups.
66 The following code excerpt illutrates a typical use of the InterpKernelDEC class.
70 InterpKernelDEC dec(groupA, groupB);
71 dec.attachLocalField(field);
73 if (groupA.containsMyRank())
75 else if (groupB.containsMyRank())
79 A \ref InterpKerRemapGlobal "remapping" of the field from the source mesh to the target mesh is performed by
80 the function synchronise(), which computes the interpolation matrix.
82 Computing the field on the receiving side can be expressed in terms of a matrix-vector product :
83 \f$ \phi_t=W.\phi_s\f$, with \f$ \phi_t \f$ the field on the target side and \f$ \phi_s \f$ the field
85 When remapping a 3D surface to another 3D surface, a projection phase is necessary to match elements
86 from both sides. Care must be taken when defining this projection to obtain a
87 \ref InterpKerRemapGlobal "conservative remapping".
89 In the P0-P0 case, this matrix is a plain rectangular matrix with coefficients equal to the
90 intersection areas between triangle and quadrangles. For instance, in the above figure, the matrix
94 \begin{tabular}{|cccc|}
96 0.46 & 0 & 0.51 & 0.03\\
97 0.42 & 0.53 & 0 & 0.05\\
98 0 & 0 & 0.92 & 0.05 \\
104 \section interpkerneldec_options Options
105 On top of \ref dec_options, options supported by %InterpKernelDEC objects are
106 related to the underlying Intersector class.
107 All the options available in the intersector objects are
108 available for the %InterpKernelDEC object. The various options available for * intersectors can
109 be reviewed in \ref InterpKerIntersectors.
113 InterpKernelDEC dec(source_group, target_group);
114 dec.attachLocalField(field);
115 dec.setOptions("DoRotate",false);
116 dec.setOptions("Precision",1e-12);
120 \warning{ Options must be set before calling the synchronize method. }
123 InterpKernelDEC::InterpKernelDEC():_interpolation_matrix(0)
128 This constructor creates an InterpKernelDEC which has \a source_group as a working side
129 and \a target_group as an idle side. All the processors will actually participate, but intersection computations will be performed on the working side during the \a synchronize() phase.
130 The constructor must be called synchronously on all processors of both processor groups.
132 \param source_group working side ProcessorGroup
133 \param target_group lazy side ProcessorGroup
136 InterpKernelDEC::InterpKernelDEC(ProcessorGroup& source_group, ProcessorGroup& target_group):
137 DisjointDEC(source_group, target_group),_interpolation_matrix(0)
142 InterpKernelDEC::InterpKernelDEC(const std::set<int>& src_ids, const std::set<int>& trg_ids,
143 const MPI_Comm& world_comm):DisjointDEC(src_ids,trg_ids,world_comm),
144 _interpolation_matrix(0)
148 InterpKernelDEC::~InterpKernelDEC()
150 if (_interpolation_matrix !=0)
151 delete _interpolation_matrix;
155 \brief Synchronization process for exchanging topologies.
157 This method prepares all the structures necessary for sending data from a processor group to the other. It uses the mesh underlying the fields that have been set with attachLocalField method.
158 It works in four steps :
159 -# Bounding boxes are computed for each subdomain,
160 -# The lazy side mesh parts that are likely to intersect the working side local processor are sent to the working side,
161 -# The working side calls the interpolation kernel to compute the intersection between local and imported mesh.
162 -# The lazy side is updated so that it knows the structure of the data that will be sent by
163 the working side during a \a sendData() call.
166 void InterpKernelDEC::synchronize()
170 delete _interpolation_matrix;
171 _interpolation_matrix = new InterpolationMatrix (_local_field, *_source_group,*_target_group,*this,*this);
173 //setting up the communication DEC on both sides
174 if (_source_group->containsMyRank())
176 //locate the distant meshes
177 ElementLocator locator(*_local_field, *_target_group, *_source_group);
178 //transfering option from InterpKernelDEC to ElementLocator
179 locator.copyOptions(*this);
180 MEDCouplingPointSet* distant_mesh=0;
182 std::string distantMeth;
183 for (int i=0; i<_target_group->size(); i++)
185 // int idistant_proc = (i+_source_group->myRank())%_target_group->size();
188 //gathers pieces of the target meshes that can intersect the local mesh
189 locator.exchangeMesh(idistant_proc,distant_mesh,distant_ids);
190 if (distant_mesh !=0)
192 locator.exchangeMethod(_method,idistant_proc,distantMeth);
193 //adds the contribution of the distant mesh on the local one
194 int idistant_proc_in_union=_union_group->translateRank(_target_group,idistant_proc);
195 //std::cout <<"add contribution from proc "<<idistant_proc_in_union<<" to proc "<<_union_group->myRank()<<std::endl;
196 _interpolation_matrix->addContribution(*distant_mesh,idistant_proc_in_union,distant_ids,_method,distantMeth);
197 distant_mesh->decrRef();
198 delete [] distant_ids;
203 _interpolation_matrix->finishContributionW(locator);
206 if (_target_group->containsMyRank())
208 ElementLocator locator(*_local_field, *_source_group, *_target_group);
209 //transfering option from InterpKernelDEC to ElementLocator
210 locator.copyOptions(*this);
211 MEDCouplingPointSet* distant_mesh=0;
213 for (int i=0; i<_source_group->size(); i++)
215 // int idistant_proc = (i+_target_group->myRank())%_source_group->size();
217 //gathers pieces of the target meshes that can intersect the local mesh
218 locator.exchangeMesh(idistant_proc,distant_mesh,distant_ids);
219 //std::cout << " Data sent from "<<_union_group->myRank()<<" to source proc "<< idistant_proc<<std::endl;
222 std::string distantMeth;
223 locator.exchangeMethod(_method,idistant_proc,distantMeth);
224 distant_mesh->decrRef();
225 delete [] distant_ids;
230 _interpolation_matrix->finishContributionL(locator);
232 _interpolation_matrix->prepare();
237 Receives the data whether the processor is on the working side or on the lazy side. It must match a \a sendData() call on the other side.
239 void InterpKernelDEC::recvData()
241 if (_source_group->containsMyRank())
242 _interpolation_matrix->transposeMultiply(*_local_field->getField());
243 else if (_target_group->containsMyRank())
245 _interpolation_matrix->multiply(*_local_field->getField());
246 if (getForcedRenormalization())
247 renormalizeTargetField(getMeasureAbsStatus());
253 Receives the data at time \a time in asynchronous mode. The value of the field
254 will be time-interpolated from the field values received.
255 \param time time at which the value is desired
257 void InterpKernelDEC::recvData( double time )
259 _interpolation_matrix->getAccessDEC()->setTime(time);
264 Sends the data whether the processor is on the working side or on the lazy side.
265 It must match a recvData() call on the other side.
267 void InterpKernelDEC::sendData()
269 if (_source_group->containsMyRank())
272 _interpolation_matrix->multiply(*_local_field->getField());
273 if (getForcedRenormalization())
274 renormalizeTargetField(getMeasureAbsStatus());
277 else if (_target_group->containsMyRank())
278 _interpolation_matrix->transposeMultiply(*_local_field->getField());
282 Sends the data available at time \a time in asynchronous mode.
283 \param time time at which the value is available
284 \param deltatime time interval between the value presently sent and the next one.
286 void InterpKernelDEC::sendData( double time , double deltatime )
288 _interpolation_matrix->getAccessDEC()->setTime(time,deltatime);