vtk-m/vtkm/filter/ContourTreeUniformDistributed.h

//============================================================================
//  Copyright (c) Kitware, Inc.
//  All rights reserved.
//  See LICENSE.txt for details.
//
//  This software is distributed WITHOUT ANY WARRANTY; without even
//  the implied warranty of MERCHANTABILITY or FITNESS FOR A PARTICULAR
//  PURPOSE.  See the above copyright notice for more information.
//============================================================================
// Copyright (c) 2018, The Regents of the University of California, through
// Lawrence Berkeley National Laboratory (subject to receipt of any required approvals
// from the U.S. Dept. of Energy).  All rights reserved.
//
// Redistribution and use in source and binary forms, with or without modification,
// are permitted provided that the following conditions are met:
//
// (1) Redistributions of source code must retain the above copyright notice, this
//     list of conditions and the following disclaimer.
//
// (2) Redistributions in binary form must reproduce the above copyright notice,
//     this list of conditions and the following disclaimer in the documentation
//     and/or other materials provided with the distribution.
//
// (3) Neither the name of the University of California, Lawrence Berkeley National
//     Laboratory, U.S. Dept. of Energy nor the names of its contributors may be
//     used to endorse or promote products derived from this software without
//     specific prior written permission.
//
// THIS SOFTWARE IS PROVIDED BY THE COPYRIGHT HOLDERS AND CONTRIBUTORS "AS IS" AND
// ANY EXPRESS OR IMPLIED WARRANTIES, INCLUDING, BUT NOT LIMITED TO, THE IMPLIED
// WARRANTIES OF MERCHANTABILITY AND FITNESS FOR A PARTICULAR PURPOSE ARE DISCLAIMED.
// IN NO EVENT SHALL THE COPYRIGHT OWNER OR CONTRIBUTORS BE LIABLE FOR ANY DIRECT,
// INDIRECT, INCIDENTAL, SPECIAL, EXEMPLARY, OR CONSEQUENTIAL DAMAGES (INCLUDING,
// BUT NOT LIMITED TO, PROCUREMENT OF SUBSTITUTE GOODS OR SERVICES; LOSS OF USE,
// DATA, OR PROFITS; OR BUSINESS INTERRUPTION) HOWEVER CAUSED AND ON ANY THEORY OF
// LIABILITY, WHETHER IN CONTRACT, STRICT LIABILITY, OR TORT (INCLUDING NEGLIGENCE
// OR OTHERWISE) ARISING IN ANY WAY OUT OF THE USE OF THIS SOFTWARE, EVEN IF ADVISED
// OF THE POSSIBILITY OF SUCH DAMAGE.
//
//=============================================================================
//
//  This code is an extension of the algorithm presented in the paper:
//  Parallel Peak Pruning for Scalable SMP Contour Tree Computation.
//  Hamish Carr, Gunther Weber, Christopher Sewell, and James Ahrens.
//  Proceedings of the IEEE Symposium on Large Data Analysis and Visualization
//  (LDAV), October 2016, Baltimore, Maryland.
//
//  The PPP2 algorithm and software were jointly developed by
//  Hamish Carr (University of Leeds), Gunther H. Weber (LBNL), and
//  Oliver Ruebel (LBNL)
//==============================================================================


#ifndef vtk_m_filter_ContourTreeUniformDistributed_h
#define vtk_m_filter_ContourTreeUniformDistributed_h

#include <vtkm/Types.h>
#include <vtkm/cont/ArrayHandle.h>
#include <vtkm/worklet/contourtree_augmented/ContourTree.h>
#include <vtkm/worklet/contourtree_augmented/DataSetMesh.h>
#include <vtkm/worklet/contourtree_distributed/BoundaryTree.h>
#include <vtkm/worklet/contourtree_distributed/HierarchicalContourTree.h>
#include <vtkm/worklet/contourtree_distributed/InteriorForest.h>
#include <vtkm/worklet/contourtree_distributed/SpatialDecomposition.h>

#include <vtkm/filter/FilterField.h>

#include <memory>

namespace vtkm
{
namespace filter
{

/// \brief Construct the Contour Tree for a 2D or 3D regular mesh
///
/// This filter implements the parallel peak pruning algorithm. In contrast to
/// the ContourTreeUniform filter, this filter is optimized to allow for the
/// computation of the augmented contour tree, i.e., the contour tree including
/// all regular mesh vertices. Augmentation with regular vertices is used in
/// practice to compute statistics (e.g., volume), to segment the input mesh,
/// facilitate iso-value selection, enable localization of all verticies of a
/// mesh in the tree among others.
///
/// In addition to single-block computation, the filter also supports multi-block
/// regular grids. The blocks are processed in parallel using DIY and then the
/// tree are merged progressively using a binary-reduction scheme to compute the
/// final contour tree. I.e., in the multi-block context, the final tree is
/// constructed on rank 0.
class ContourTreeUniformDistributed
  : public vtkm::filter::FilterField<ContourTreeUniformDistributed>
{
public:
  using SupportedTypes = vtkm::TypeListScalarAll;
  ///
  /// Create the contour tree filter
  /// @param[in] blocksPerDim  Number of data blocks used in each data dimension
  /// @param[in] globalSize  Global extends of the input mesh (i.e., number of mesh points in each dimension)
  /// @param[in] localBlockIndices  Array with the (x,y,z) index of each local data block with
  ///                              with respect to blocksPerDim
  /// @param[in] localBlockOrigins  Array with the (x,y,z) origin (with regard to mesh index) of each
  ///                              local data block
  /// @param[in] localBlockSizes    Array with the sizes (i.e., extends in number of mesh points) of each
  ///                             local data block
  /// @param[in] useMarchingCubes Boolean indicating whether marching cubes (true) or freudenthal (false)
  ///                            connectivity should be used. Valid only for 3D input data. Default is false.
  /// @param[in] saveDotFiles Save debug dot output files for the distributed contour tree compute.
  /// @param[in] timingsLogLevel Set the vtkm::cont:LogLevel to be used to record timings information
  ///                            specific to the computation of the hierachical contour tree
  /// @param[in] treeLogLevel Set the vtkm::cont:LogLevel to be used to record metadata information
  ///                         about the various trees computed as part of the hierarchical contour tree compute
  VTKM_CONT
  ContourTreeUniformDistributed(
    vtkm::Id3 blocksPerDim, // TODO/FIXME: Possibly pass SpatialDecomposition object instead
    vtkm::Id3 globalSize,
    const vtkm::cont::ArrayHandle<vtkm::Id3>& localBlockIndices,
    const vtkm::cont::ArrayHandle<vtkm::Id3>& localBlockOrigins,
    const vtkm::cont::ArrayHandle<vtkm::Id3>& localBlockSizes,
    bool useBoundaryExtremaOnly = true,
    bool useMarchingCubes = false,
    bool saveDotFiles = false,
    vtkm::cont::LogLevel timingsLogLevel = vtkm::cont::LogLevel::Perf,
    vtkm::cont::LogLevel treeLogLevel = vtkm::cont::LogLevel::Info);

  template <typename DerivedPolicy>
  VTKM_CONT vtkm::cont::PartitionedDataSet PrepareForExecution(
    const vtkm::cont::PartitionedDataSet& input,
    const vtkm::filter::PolicyBase<DerivedPolicy>& policy);

  template <typename T, typename StorageType, typename DerivedPolicy>
  VTKM_CONT void ComputeLocalTree(const vtkm::Id blockIndex,
                                  const vtkm::cont::DataSet& input,
                                  const vtkm::cont::ArrayHandle<T, StorageType>& field,
                                  const vtkm::filter::FieldMetadata& fieldMeta,
                                  vtkm::filter::PolicyBase<DerivedPolicy> policy);

  /// Implement per block contour tree computation after the MeshType has been discovered
  template <typename T,
            typename StorageType,
            typename DerivedPolicy,
            typename MeshType,
            typename MeshBoundaryExecType>
  VTKM_CONT void ComputeLocalTreeImpl(const vtkm::Id blockIndex,
                                      const vtkm::cont::DataSet& input,
                                      const vtkm::cont::ArrayHandle<T, StorageType>& field,
                                      const vtkm::filter::FieldMetadata& fieldMeta,
                                      vtkm::filter::PolicyBase<DerivedPolicy> policy,
                                      MeshType& mesh,
                                      MeshBoundaryExecType& meshBoundaryExecObject);

  //@{
  /// when operating on vtkm::cont::MultiBlock we want to
  /// do processing across ranks as well. Just adding pre/post handles
  /// for the same does the trick.
  template <typename DerivedPolicy>
  VTKM_CONT void PreExecute(const vtkm::cont::PartitionedDataSet& input,
                            const vtkm::filter::PolicyBase<DerivedPolicy>& policy);

  template <typename DerivedPolicy>
  VTKM_CONT void PostExecute(const vtkm::cont::PartitionedDataSet& input,
                             vtkm::cont::PartitionedDataSet& output,
                             const vtkm::filter::PolicyBase<DerivedPolicy>&);


  ///
  /// Internal helper function that implements the actual functionality of PostExecute
  ///
  /// In the case we operate on vtkm::cont::MultiBlock we need to merge the trees
  /// computed on the block to compute the final contour tree.
  template <typename T, typename StorageType, typename DerivedPolicy>
  VTKM_CONT void DoPostExecute(
    const vtkm::cont::PartitionedDataSet& input,
    vtkm::cont::PartitionedDataSet& output,
    const vtkm::filter::FieldMetadata& fieldMeta,
    const vtkm::cont::ArrayHandle<T, StorageType>&, // dummy parameter to get the type
    vtkm::filter::PolicyBase<DerivedPolicy> policy);
  //@}

private:
  /// Use only boundary critical points in the parallel merge to reduce communication. Disabling this should only be needed for performance testing.
  bool UseBoundaryExtremaOnly;

  /// Use marching cubes connectivity for computing the contour tree
  bool UseMarchingCubes;

  /// Save dot files for all tree computations
  bool SaveDotFiles;

  /// Log level to be used for outputting timing information. Default is vtkm::cont::LogLevel::Perf
  vtkm::cont::LogLevel TimingsLogLevel = vtkm::cont::LogLevel::Perf;

  /// Log level to be used for outputting metadata about the trees. Default is vtkm::cont::LogLevel::Info
  vtkm::cont::LogLevel TreeLogLevel = vtkm::cont::LogLevel::Info;

  /// Information about the spatial decomposition
  vtkm::worklet::contourtree_distributed::SpatialDecomposition MultiBlockSpatialDecomposition;

  /// Intermediate results (one per local data block)...
  /// ... local mesh information needed at end of fan out
  std::vector<vtkm::worklet::contourtree_augmented::DataSetMesh> LocalMeshes;
  /// ... local contour trees etc. computed during fan in and used during fan out
  std::vector<vtkm::worklet::contourtree_augmented::ContourTree> LocalContourTrees;
  std::vector<vtkm::worklet::contourtree_distributed::BoundaryTree> LocalBoundaryTrees;
  std::vector<vtkm::worklet::contourtree_distributed::InteriorForest> LocalInteriorForests;

  /// The hierarchical trees computed by the filter (array with one entry per block)
  // TODO/FIXME: We need to find a way to store the final hieararchical trees somewhere.
  // Currently we cannot do this here as it is a template on FieldType
  //
  //std::vector<vtkm::worklet::contourtree_distributed::HierarchicalContourTree> HierarchicalContourTrees;
  /// Number of iterations used to compute the contour tree
  vtkm::Id NumIterations;
};

} // namespace filter
} // namespace vtkm

#include <vtkm/filter/ContourTreeUniformDistributed.hxx>

#endif // vtk_m_filter_ContourTreeUniformAugmented_h