libMesh::SubdomainPartitioner Class Reference

Independently partitions chunks of subdomains and combines the results. More...

#include <subdomain_partitioner.h>

Inheritance diagram for libMesh::SubdomainPartitioner:

Public Member Functions
	SubdomainPartitioner ()
	SubdomainPartitioner (const SubdomainPartitioner &other)
SubdomainPartitioner &	operator= (const SubdomainPartitioner &)=delete
	SubdomainPartitioner (SubdomainPartitioner &&)=default
SubdomainPartitioner &	operator= (SubdomainPartitioner &&)=default
virtual	~SubdomainPartitioner ()=default
virtual std::unique_ptr< Partitioner >	clone () const override
std::unique_ptr< Partitioner > &	internal_partitioner ()
virtual void	partition (MeshBase &mesh, const unsigned int n)
virtual void	partition (MeshBase &mesh)
virtual void	partition_range (MeshBase &, MeshBase::element_iterator, MeshBase::element_iterator, const unsigned int)
void	repartition (MeshBase &mesh, const unsigned int n)
void	repartition (MeshBase &mesh)
virtual void	attach_weights (ErrorVector *)

Static Public Member Functions
static void	partition_unpartitioned_elements (MeshBase &mesh)
static void	partition_unpartitioned_elements (MeshBase &mesh, const unsigned int n)
static void	set_parent_processor_ids (MeshBase &mesh)
static void	set_node_processor_ids (MeshBase &mesh)
static void	processor_pairs_to_interface_nodes (MeshBase &mesh, std::map< std::pair< processor_id_type, processor_id_type >, std::set< dof_id_type >> &processor_pair_to_nodes)
static void	set_interface_node_processor_ids_linear (MeshBase &mesh)
static void	set_interface_node_processor_ids_BFS (MeshBase &mesh)
static void	set_interface_node_processor_ids_petscpartitioner (MeshBase &mesh)

Public Attributes
std::vector< std::set< subdomain_id_type > >	chunks

Protected Member Functions
virtual void	_do_partition (MeshBase &mesh, const unsigned int n) override
void	single_partition (MeshBase &mesh)
void	single_partition_range (MeshBase::element_iterator it, MeshBase::element_iterator end)
virtual void	_do_repartition (MeshBase &mesh, const unsigned int n)
virtual void	_find_global_index_by_pid_map (const MeshBase &mesh)
virtual void	build_graph (const MeshBase &mesh)
void	assign_partitioning (const MeshBase &mesh, const std::vector< dof_id_type > &parts)

Protected Attributes
std::unique_ptr< Partitioner >	_internal_partitioner
ErrorVector *	_weights
std::unordered_map< dof_id_type, dof_id_type >	_global_index_by_pid_map
std::vector< dof_id_type >	_n_active_elem_on_proc
std::vector< std::vector< dof_id_type > >	_dual_graph
std::vector< Elem * >	_local_id_to_elem

Static Protected Attributes
static const dof_id_type	communication_blocksize = 1000000

Detailed Description

Independently partitions chunks of subdomains and combines the results.

The SubdomainPartitioner partitions the elements in "chunks" of user-specified subdomain ids. Once all the chunks are partitioned, the overall mesh partitioning is simply the union of the chunk partitionings. For example, if the "chunks" vector is given by: chunks[0] = {1, 2, 4} chunks[1] = {3, 7, 8} chunks[2] = {5, 6} then we will call the internal Partitioner three times, once for subdomains 1, 2, and 4, once for subdomains 3, 7, and 8, and once for subdomains 5 and 6.

Note

This Partitioner may produce highly non-optimal communication patterns and is likely to place geometrically disjoint sets of elements on the same processor. Its intended use is to help facilitate load balancing. That is, if the user knows that certain subdomains (or groups of subdomains) are more expensive to compute than others, he/she can ensure that they are partitioned more or less evenly among the available processors by specifying them together in a single entry of the "chunk" vector.

Author

John W. Peterson

Date

2017

Definition at line 55 of file subdomain_partitioner.h.

Constructor & Destructor Documentation

SubdomainPartitioner() [1/3]

libMesh::SubdomainPartitioner::SubdomainPartitioner

(

)

Constructors. The default ctor initializes the internal Partitioner object to a MetisPartitioner so the class is usable, although this type can be customized later.

Definition at line 29 of file subdomain_partitioner.C.

                                            :
  _internal_partitioner(libmesh_make_unique<MetisPartitioner>())
{}

SubdomainPartitioner() [2/3]

libMesh::SubdomainPartitioner::SubdomainPartitioner

(

const SubdomainPartitioner &

other

)

Definition at line 34 of file subdomain_partitioner.C.

  : Partitioner(other),
    chunks(other.chunks),
    _internal_partitioner(other._internal_partitioner->clone())
{}

SubdomainPartitioner() [3/3]

libMesh::SubdomainPartitioner::SubdomainPartitioner ( SubdomainPartitioner &&  )

default

Move ctor, move assignment operator, and destructor are all explicitly defaulted for this class.

~SubdomainPartitioner()

virtual libMesh::SubdomainPartitioner::~SubdomainPartitioner ( )

virtualdefault

Member Function Documentation

_do_partition()

void libMesh::SubdomainPartitioner::_do_partition ( MeshBase &  mesh, const unsigned int  n  )

overrideprotectedvirtual

Partition the MeshBase into n subdomains.

Implements libMesh::Partitioner.

Definition at line 41 of file subdomain_partitioner.C.

References _internal_partitioner, chunks, end, mesh, and libMesh::Partitioner::single_partition().

{
  libmesh_assert_greater (n, 0);

  // Check for an easy return.  If the user has not specified any
  // entries in the chunks vector, we just do a single partitioning.
  if ((n == 1) || chunks.empty())
    {
      this->single_partition (mesh);
      return;
    }

  // Now actually do the partitioning.
  LOG_SCOPE ("_do_partition()", "SubdomainPartitioner");

  // For each chunk, construct an iterator range for the set of
  // subdomains in question, and pass it to the internal Partitioner.
  for (std::size_t c=0; c<chunks.size(); ++c)
    {
      MeshBase::element_iterator
        it = mesh.active_subdomain_set_elements_begin(chunks[c]),
        end = mesh.active_subdomain_set_elements_end(chunks[c]);

      _internal_partitioner->partition_range(mesh, it, end, n);
    }
}

_do_repartition()

virtual void libMesh::Partitioner::_do_repartition ( MeshBase &  mesh, const unsigned int  n  )

inlineprotectedvirtualinherited

This is the actual re-partitioning method which can be overridden in derived classes.

Note

The default behavior is to simply call the partition function.

Reimplemented in libMesh::ParmetisPartitioner.

Definition at line 237 of file partitioner.h.

References libMesh::Partitioner::_do_partition().

Referenced by libMesh::Partitioner::repartition().

                                                      { this->_do_partition (mesh, n); }

_find_global_index_by_pid_map()

void libMesh::Partitioner::_find_global_index_by_pid_map ( const MeshBase &  mesh )

protectedvirtualinherited

Construct contiguous global indices for the current partitioning. The global indices are ordered part-by-part

Definition at line 907 of file partitioner.C.

References libMesh::Partitioner::_global_index_by_pid_map, libMesh::Partitioner::_n_active_elem_on_proc, libMesh::as_range(), libMesh::MeshTools::create_bounding_box(), libMesh::MeshCommunication::find_local_indices(), mesh, and libMesh::Parallel::sync_dofobject_data_by_id().

Referenced by libMesh::Partitioner::build_graph().

{
  const dof_id_type n_active_local_elem = mesh.n_active_local_elem();

  // Find the number of active elements on each processor.  We cannot use
  // mesh.n_active_elem_on_proc(pid) since that only returns the number of
  // elements assigned to pid which are currently stored on the calling
  // processor. This will not in general be correct for parallel meshes
  // when (pid!=mesh.processor_id()).
  _n_active_elem_on_proc.resize(mesh.n_processors());
  mesh.comm().allgather(n_active_local_elem, _n_active_elem_on_proc);

  libMesh::BoundingBox bbox =
    MeshTools::create_bounding_box(mesh);

  _global_index_by_pid_map.clear();

  // create the mapping which is contiguous by processor
  MeshCommunication().find_local_indices (bbox,
                                          mesh.active_local_elements_begin(),
                                          mesh.active_local_elements_end(),
                                          _global_index_by_pid_map);

  SyncLocalIDs sync(_global_index_by_pid_map);

  Parallel::sync_dofobject_data_by_id
      (mesh.comm(), mesh.active_elements_begin(), mesh.active_elements_end(), sync);

  dof_id_type pid_offset=0;
  for (processor_id_type pid=0; pid<mesh.n_processors(); pid++)
    {
      for (const auto & elem : as_range(mesh.active_pid_elements_begin(pid),
                                        mesh.active_pid_elements_end(pid)))
        {
          libmesh_assert_less (_global_index_by_pid_map[elem->id()], _n_active_elem_on_proc[pid]);

          _global_index_by_pid_map[elem->id()] += pid_offset;
        }

      pid_offset += _n_active_elem_on_proc[pid];
    }
}

assign_partitioning()

void libMesh::Partitioner::assign_partitioning ( const MeshBase &  mesh, const std::vector< dof_id_type > &  parts  )

protectedinherited

Assign the computed partitioning to the mesh.

Definition at line 1113 of file partitioner.C.

References libMesh::Partitioner::_global_index_by_pid_map, libMesh::Partitioner::_n_active_elem_on_proc, data, mesh, and libMesh::Parallel::pull_parallel_vector_data().

{
  LOG_SCOPE("assign_partitioning()", "ParmetisPartitioner");

  // This function must be run on all processors at once
  libmesh_parallel_only(mesh.comm());

  dof_id_type first_local_elem = 0;
  for (processor_id_type pid=0; pid < mesh.processor_id(); pid++)
    first_local_elem += _n_active_elem_on_proc[pid];

#ifndef NDEBUG
  const dof_id_type n_active_local_elem = mesh.n_active_local_elem();
#endif

  std::map<processor_id_type, std::vector<dof_id_type>>
    requested_ids;

  // Results to gather from each processor - kept in a map so we
  // do only one loop over elements after all receives are done.
  std::map<processor_id_type, std::vector<processor_id_type>>
    filled_request;

  for (auto & elem : mesh.active_element_ptr_range())
    {
      // we need to get the index from the owning processor
      // (note we cannot assign it now -- we are iterating
      // over elements again and this will be bad!)
      requested_ids[elem->processor_id()].push_back(elem->id());
    }

  auto gather_functor =
    [this,
     & parts,
#ifndef NDEBUG
     & mesh,
     n_active_local_elem,
#endif
     first_local_elem]
    (processor_id_type, const std::vector<dof_id_type> & ids,
     std::vector<processor_id_type> & data)
    {
      const std::size_t ids_size = ids.size();
      data.resize(ids.size());

      for (std::size_t i=0; i != ids_size; i++)
        {
          const dof_id_type requested_elem_index = ids[i];

          libmesh_assert(_global_index_by_pid_map.count(requested_elem_index));

          const dof_id_type global_index_by_pid =
            _global_index_by_pid_map[requested_elem_index];

          const dof_id_type local_index =
            global_index_by_pid - first_local_elem;

          libmesh_assert_less (local_index, parts.size());
          libmesh_assert_less (local_index, n_active_local_elem);

          const processor_id_type elem_procid =
            cast_int<processor_id_type>(parts[local_index]);

          libmesh_assert_less (elem_procid, mesh.n_partitions());

          data[i] = elem_procid;
        }
    };

  auto action_functor =
    [&filled_request]
    (processor_id_type pid,
     const std::vector<dof_id_type> &,
     const std::vector<processor_id_type> & new_procids)
    {
      filled_request[pid] = new_procids;
    };

  // Trade requests with other processors
  const processor_id_type * ex = nullptr;
  Parallel::pull_parallel_vector_data
    (mesh.comm(), requested_ids, gather_functor, action_functor, ex);

  // and finally assign the partitioning.
  // note we are iterating in exactly the same order
  // used to build up the request, so we can expect the
  // required entries to be in the proper sequence.
  std::vector<unsigned int> counters(mesh.n_processors(), 0);
  for (auto & elem : mesh.active_element_ptr_range())
    {
      const processor_id_type current_pid = elem->processor_id();

      libmesh_assert_less (counters[current_pid], requested_ids[current_pid].size());

      const processor_id_type elem_procid =
        filled_request[current_pid][counters[current_pid]++];

      libmesh_assert_less (elem_procid, mesh.n_partitions());
      elem->processor_id() = elem_procid;
    }
}

attach_weights()

virtual void libMesh::Partitioner::attach_weights ( ErrorVector *  )

inlinevirtualinherited

Attach weights that can be used for partitioning. This ErrorVector should be exactly the same on every processor and should have mesh->max_elem_id() entries.

Reimplemented in libMesh::MetisPartitioner.

Definition at line 203 of file partitioner.h.

{ libmesh_not_implemented(); }

build_graph()

void libMesh::Partitioner::build_graph ( const MeshBase &  mesh )

protectedvirtualinherited

Build a dual graph for partitioner

Reimplemented in libMesh::ParmetisPartitioner.

Definition at line 950 of file partitioner.C.

References libMesh::Partitioner::_dual_graph, libMesh::Partitioner::_find_global_index_by_pid_map(), libMesh::Partitioner::_global_index_by_pid_map, libMesh::Partitioner::_local_id_to_elem, libMesh::Partitioner::_n_active_elem_on_proc, libMesh::Elem::active(), libMesh::as_range(), libMesh::DofObject::id(), mesh, and libMesh::Elem::neighbor_ptr().

{
  LOG_SCOPE("build_graph()", "ParmetisPartitioner");

  const dof_id_type n_active_local_elem  = mesh.n_active_local_elem();
  // If we have boundary elements in this mesh, we want to account for
  // the connectivity between them and interior elements.  We can find
  // interior elements from boundary elements, but we need to build up
  // a lookup map to do the reverse.
  typedef std::unordered_multimap<const Elem *, const Elem *> map_type;
  map_type interior_to_boundary_map;

  for (const auto & elem : mesh.active_element_ptr_range())
    {
      // If we don't have an interior_parent then there's nothing to look us
      // up.
      if ((elem->dim() >= LIBMESH_DIM) ||
          !elem->interior_parent())
        continue;

      // get all relevant interior elements
      std::set<const Elem *> neighbor_set;
      elem->find_interior_neighbors(neighbor_set);

      for (const auto & neighbor : neighbor_set)
        interior_to_boundary_map.insert(std::make_pair(neighbor, elem));
    }

#ifdef LIBMESH_ENABLE_AMR
  std::vector<const Elem *> neighbors_offspring;
#endif

   // This is costly, and we only need to do it if the mesh has
   // changed since we last partitioned... but the mesh probably has
   // changed since we last partitioned, and if it hasn't we don't
   // have a reliable way to be sure of that.
   _find_global_index_by_pid_map(mesh);

   dof_id_type first_local_elem = 0;
   for (processor_id_type pid=0; pid < mesh.processor_id(); pid++)
     first_local_elem += _n_active_elem_on_proc[pid];

  _dual_graph.clear();
  _dual_graph.resize(n_active_local_elem);
  _local_id_to_elem.resize(n_active_local_elem);

  for (const auto & elem : mesh.active_local_element_ptr_range())
    {
      libmesh_assert (_global_index_by_pid_map.count(elem->id()));
      const dof_id_type global_index_by_pid =
        _global_index_by_pid_map[elem->id()];

      const dof_id_type local_index =
        global_index_by_pid - first_local_elem;
      libmesh_assert_less (local_index, n_active_local_elem);

      std::vector<dof_id_type> & graph_row = _dual_graph[local_index];

      // Save this off to make it easy to index later
      _local_id_to_elem[local_index] = elem;

      // Loop over the element's neighbors.  An element
      // adjacency corresponds to a face neighbor
      for (auto neighbor : elem->neighbor_ptr_range())
        {
          if (neighbor != nullptr)
            {
              // If the neighbor is active treat it
              // as a connection
              if (neighbor->active())
                {
                  libmesh_assert(_global_index_by_pid_map.count(neighbor->id()));
                  const dof_id_type neighbor_global_index_by_pid =
                    _global_index_by_pid_map[neighbor->id()];

                  graph_row.push_back(neighbor_global_index_by_pid);
                }

#ifdef LIBMESH_ENABLE_AMR

              // Otherwise we need to find all of the
              // neighbor's children that are connected to
              // us and add them
              else
                {
                  // The side of the neighbor to which
                  // we are connected
                  const unsigned int ns =
                    neighbor->which_neighbor_am_i (elem);
                  libmesh_assert_less (ns, neighbor->n_neighbors());

                  // Get all the active children (& grandchildren, etc...)
                  // of the neighbor

                  // FIXME - this is the wrong thing, since we
                  // should be getting the active family tree on
                  // our side only.  But adding too many graph
                  // links may cause hanging nodes to tend to be
                  // on partition interiors, which would reduce
                  // communication overhead for constraint
                  // equations, so we'll leave it.

                  neighbor->active_family_tree (neighbors_offspring);

                  // Get all the neighbor's children that
                  // live on that side and are thus connected
                  // to us
                  for (std::size_t nc=0; nc<neighbors_offspring.size(); nc++)
                    {
                      const Elem * child =
                        neighbors_offspring[nc];

                      // This does not assume a level-1 mesh.
                      // Note that since children have sides numbered
                      // coincident with the parent then this is a sufficient test.
                      if (child->neighbor_ptr(ns) == elem)
                        {
                          libmesh_assert (child->active());
                          libmesh_assert (_global_index_by_pid_map.count(child->id()));
                          const dof_id_type child_global_index_by_pid =
                            _global_index_by_pid_map[child->id()];

                          graph_row.push_back(child_global_index_by_pid);
                        }
                    }
                }

#endif /* ifdef LIBMESH_ENABLE_AMR */


            }
        }

      if ((elem->dim() < LIBMESH_DIM) &&
          elem->interior_parent())
        {
          // get all relevant interior elements
          std::set<const Elem *> neighbor_set;
          elem->find_interior_neighbors(neighbor_set);

          for (const auto & neighbor : neighbor_set)
            {
              const dof_id_type neighbor_global_index_by_pid =
                _global_index_by_pid_map[neighbor->id()];

              graph_row.push_back(neighbor_global_index_by_pid);
            }
        }

      // Check for any boundary neighbors
      for (const auto & pr : as_range(interior_to_boundary_map.equal_range(elem)))
        {
          const Elem * neighbor = pr.second;

          const dof_id_type neighbor_global_index_by_pid =
            _global_index_by_pid_map[neighbor->id()];

          graph_row.push_back(neighbor_global_index_by_pid);
        }
    }

}

clone()

virtual std::unique_ptr<Partitioner> libMesh::SubdomainPartitioner::clone ( ) const

inlineoverridevirtual

Returns

A copy of this partitioner wrapped in a smart pointer.

Implements libMesh::Partitioner.

Definition at line 84 of file subdomain_partitioner.h.

  {
    return libmesh_make_unique<SubdomainPartitioner>(*this);
  }

internal_partitioner()

std::unique_ptr<Partitioner>& libMesh::SubdomainPartitioner::internal_partitioner ( )

inline

Get a reference to the Partitioner used internally by the SubdomainPartitioner.

Note

The internal Partitioner cannot also be a SubdomainPartitioner, otherwise an infinite loop will result. To have this class use e.g. the space-filling curve Partitioner internally, one could do:

SubdomainPartitioner sp;
sp.internal_partitioner().reset(new SFCPartitioner);

Definition at line 111 of file subdomain_partitioner.h.

References _internal_partitioner.

{ return _internal_partitioner; }

operator=() [1/2]

SubdomainPartitioner& libMesh::SubdomainPartitioner::operator= ( const SubdomainPartitioner &  )

delete

This class contains a unique_ptr member, so it can't be default copy assigned.

operator=() [2/2]

SubdomainPartitioner& libMesh::SubdomainPartitioner::operator= ( SubdomainPartitioner &&  )

default

partition() [1/2]

void libMesh::Partitioner::partition ( MeshBase &  mesh, const unsigned int  n  )

virtualinherited

Partitions the MeshBase into n parts by setting processor_id() on Nodes and Elems.

Note

If you are implementing a new type of Partitioner, you most likely do not want to override the partition() function, see instead the protected virtual _do_partition() method below. The partition() function is responsible for doing a lot of libmesh-internals-specific setup and finalization before and after the _do_partition() function is called. The only responsibility of the _do_partition() function, on the other hand, is to set the processor IDs of the elements according to a specific partitioning algorithm. See, e.g. MetisPartitioner for an example.

Definition at line 57 of file partitioner.C.

References libMesh::Partitioner::_do_partition(), libMesh::MeshTools::libmesh_assert_valid_remote_elems(), mesh, std::min(), libMesh::Partitioner::partition_unpartitioned_elements(), libMesh::Partitioner::set_node_processor_ids(), libMesh::Partitioner::set_parent_processor_ids(), and libMesh::Partitioner::single_partition().

Referenced by libMesh::ParmetisPartitioner::_do_repartition(), and libMesh::Partitioner::partition().

{
  libmesh_parallel_only(mesh.comm());

  // BSK - temporary fix while redistribution is integrated 6/26/2008
  // Uncomment this to not repartition in parallel
  //   if (!mesh.is_serial())
  //     return;

  // we cannot partition into more pieces than we have
  // active elements!
  const unsigned int n_parts =
    static_cast<unsigned int>
    (std::min(mesh.n_active_elem(), static_cast<dof_id_type>(n)));

  // Set the number of partitions in the mesh
  mesh.set_n_partitions()=n_parts;

  if (n_parts == 1)
    {
      this->single_partition (mesh);
      return;
    }

  // First assign a temporary partitioning to any unpartitioned elements
  Partitioner::partition_unpartitioned_elements(mesh, n_parts);

  // Call the partitioning function
  this->_do_partition(mesh,n_parts);

  // Set the parent's processor ids
  Partitioner::set_parent_processor_ids(mesh);

  // Redistribute elements if necessary, before setting node processor
  // ids, to make sure those will be set consistently
  mesh.redistribute();

#ifdef DEBUG
  MeshTools::libmesh_assert_valid_remote_elems(mesh);

  // Messed up elem processor_id()s can leave us without the child
  // elements we need to restrict vectors on a distributed mesh
  MeshTools::libmesh_assert_valid_procids<Elem>(mesh);
#endif

  // Set the node's processor ids
  Partitioner::set_node_processor_ids(mesh);

#ifdef DEBUG
  MeshTools::libmesh_assert_valid_procids<Elem>(mesh);
#endif

  // Give derived Mesh classes a chance to update any cached data to
  // reflect the new partitioning
  mesh.update_post_partitioning();
}

partition() [2/2]

void libMesh::Partitioner::partition ( MeshBase &  mesh )

virtualinherited

Partitions the MeshBase into mesh.n_processors() by setting processor_id() on Nodes and Elems.

Note

If you are implementing a new type of Partitioner, you most likely do not want to override the partition() function, see instead the protected virtual _do_partition() method below. The partition() function is responsible for doing a lot of libmesh-internals-specific setup and finalization before and after the _do_partition() function is called. The only responsibility of the _do_partition() function, on the other hand, is to set the processor IDs of the elements according to a specific partitioning algorithm. See, e.g. MetisPartitioner for an example.

Definition at line 50 of file partitioner.C.

References mesh, and libMesh::Partitioner::partition().

{
  this->partition(mesh,mesh.n_processors());
}

partition_range()

virtual void libMesh::Partitioner::partition_range ( MeshBase &  , MeshBase::element_iterator  , MeshBase::element_iterator  , const unsigned int    )

inlinevirtualinherited

Partitions elements in the range (it, end) into n parts. The mesh from which the iterators are created must also be passed in, since it is a parallel object and has other useful information in it.

Although partition_range() is part of the public Partitioner interface, it should not generally be called by applications. Its main purpose is to support the SubdomainPartitioner, which uses it internally to individually partition ranges of elements before combining them into the final partitioning. Most of the time, the protected _do_partition() function is implemented in terms of partition_range() by passing a range which includes all the elements of the Mesh.

Reimplemented in libMesh::CentroidPartitioner, libMesh::SFCPartitioner, libMesh::MappedSubdomainPartitioner, libMesh::LinearPartitioner, and libMesh::MetisPartitioner.

Definition at line 127 of file partitioner.h.

  { libmesh_not_implemented(); }

partition_unpartitioned_elements() [1/2]

void libMesh::Partitioner::partition_unpartitioned_elements ( MeshBase &  mesh )

staticinherited

These functions assign processor IDs to newly-created elements (in parallel) which are currently assigned to processor 0.

Definition at line 187 of file partitioner.C.

References mesh.

Referenced by libMesh::Partitioner::partition(), and libMesh::Partitioner::repartition().

{
  Partitioner::partition_unpartitioned_elements(mesh, mesh.n_processors());
}

partition_unpartitioned_elements() [2/2]

void libMesh::Partitioner::partition_unpartitioned_elements ( MeshBase &  mesh, const unsigned int  n  )

staticinherited

Definition at line 194 of file partitioner.C.

References libMesh::as_range(), libMesh::MeshTools::create_bounding_box(), end, libMesh::MeshCommunication::find_global_indices(), mesh, and libMesh::MeshTools::n_elem().

{
  MeshBase::element_iterator       it  = mesh.unpartitioned_elements_begin();
  const MeshBase::element_iterator end = mesh.unpartitioned_elements_end();

  const dof_id_type n_unpartitioned_elements = MeshTools::n_elem (it, end);

  // the unpartitioned elements must exist on all processors. If the range is empty on one
  // it is empty on all, and we can quit right here.
  if (!n_unpartitioned_elements)
    return;

  // find the target subdomain sizes
  std::vector<dof_id_type> subdomain_bounds(mesh.n_processors());

  for (processor_id_type pid=0; pid<mesh.n_processors(); pid++)
    {
      dof_id_type tgt_subdomain_size = 0;

      // watch out for the case that n_subdomains < n_processors
      if (pid < n_subdomains)
        {
          tgt_subdomain_size = n_unpartitioned_elements/n_subdomains;

          if (pid < n_unpartitioned_elements%n_subdomains)
            tgt_subdomain_size++;

        }

      //libMesh::out << "pid, #= " << pid << ", " << tgt_subdomain_size << std::endl;
      if (pid == 0)
        subdomain_bounds[0] = tgt_subdomain_size;
      else
        subdomain_bounds[pid] = subdomain_bounds[pid-1] + tgt_subdomain_size;
    }

  libmesh_assert_equal_to (subdomain_bounds.back(), n_unpartitioned_elements);

  // create the unique mapping for all unpartitioned elements independent of partitioning
  // determine the global indexing for all the unpartitioned elements
  std::vector<dof_id_type> global_indices;

  // Calling this on all processors a unique range in [0,n_unpartitioned_elements) is constructed.
  // Only the indices for the elements we pass in are returned in the array.
  MeshCommunication().find_global_indices (mesh.comm(),
                                           MeshTools::create_bounding_box(mesh), it, end,
                                           global_indices);

  dof_id_type cnt=0;
  for (auto & elem : as_range(it, end))
    {
      libmesh_assert_less (cnt, global_indices.size());
      const dof_id_type global_index =
        global_indices[cnt++];

      libmesh_assert_less (global_index, subdomain_bounds.back());
      libmesh_assert_less (global_index, n_unpartitioned_elements);

      const processor_id_type subdomain_id =
        cast_int<processor_id_type>
        (std::distance(subdomain_bounds.begin(),
                       std::upper_bound(subdomain_bounds.begin(),
                                        subdomain_bounds.end(),
                                        global_index)));
      libmesh_assert_less (subdomain_id, n_subdomains);

      elem->processor_id() = subdomain_id;
      //libMesh::out << "assigning " << global_index << " to " << subdomain_id << std::endl;
    }
}

processor_pairs_to_interface_nodes()

void libMesh::Partitioner::processor_pairs_to_interface_nodes ( MeshBase &  mesh, std::map< std::pair< processor_id_type, processor_id_type >, std::set< dof_id_type >> &  processor_pair_to_nodes  )

staticinherited

On the partitioning interface, a surface is shared by two and only two processors. Try to find which pair of processors corresponds to which surfaces, and store their nodes.

Definition at line 421 of file partitioner.C.

References libMesh::DofObject::invalid_processor_id, std::max(), mesh, std::min(), and n_nodes.

Referenced by libMesh::Partitioner::set_interface_node_processor_ids_BFS(), libMesh::Partitioner::set_interface_node_processor_ids_linear(), and libMesh::Partitioner::set_interface_node_processor_ids_petscpartitioner().

{
  // This function must be run on all processors at once
  libmesh_parallel_only(mesh.comm());

  processor_pair_to_nodes.clear();

  std::set<dof_id_type> mynodes;
  std::set<dof_id_type> neighbor_nodes;
  std::vector<dof_id_type> common_nodes;

  // Loop over all the active elements
  for (auto & elem : mesh.active_element_ptr_range())
    {
      libmesh_assert(elem);

      libmesh_assert_not_equal_to (elem->processor_id(), DofObject::invalid_processor_id);

      auto n_nodes = elem->n_nodes();

      // prepare data for this element
      mynodes.clear();
      neighbor_nodes.clear();
      common_nodes.clear();

      for (unsigned int inode = 0; inode < n_nodes; inode++)
        mynodes.insert(elem->node_id(inode));

      for (auto i : elem->side_index_range())
        {
          auto neigh = elem->neighbor_ptr(i);
          if (neigh && !neigh->is_remote() && neigh->processor_id() != elem->processor_id())
            {
              neighbor_nodes.clear();
              common_nodes.clear();
              auto neigh_n_nodes = neigh->n_nodes();
              for (unsigned int inode = 0; inode < neigh_n_nodes; inode++)
                neighbor_nodes.insert(neigh->node_id(inode));

              std::set_intersection(mynodes.begin(), mynodes.end(),
                                    neighbor_nodes.begin(), neighbor_nodes.end(),
                                    std::back_inserter(common_nodes));

              auto & map_set = processor_pair_to_nodes[std::make_pair(std::min(elem->processor_id(), neigh->processor_id()),
                                                                      std::max(elem->processor_id(), neigh->processor_id()))];
              for (auto global_node_id : common_nodes)
                map_set.insert(global_node_id);
            }
        }
    }
}

repartition() [1/2]

void libMesh::Partitioner::repartition ( MeshBase &  mesh, const unsigned int  n  )

inherited

Repartitions the MeshBase into n parts. (Some partitioning algorithms can repartition more efficiently than computing a new partitioning from scratch.) The default behavior is to simply call this->partition(mesh,n).

Definition at line 124 of file partitioner.C.

References libMesh::Partitioner::_do_repartition(), mesh, std::min(), libMesh::Partitioner::partition_unpartitioned_elements(), libMesh::Partitioner::set_node_processor_ids(), libMesh::Partitioner::set_parent_processor_ids(), and libMesh::Partitioner::single_partition().

Referenced by libMesh::Partitioner::repartition().

{
  // we cannot partition into more pieces than we have
  // active elements!
  const unsigned int n_parts =
    static_cast<unsigned int>
    (std::min(mesh.n_active_elem(), static_cast<dof_id_type>(n)));

  // Set the number of partitions in the mesh
  mesh.set_n_partitions()=n_parts;

  if (n_parts == 1)
    {
      this->single_partition (mesh);
      return;
    }

  // First assign a temporary partitioning to any unpartitioned elements
  Partitioner::partition_unpartitioned_elements(mesh, n_parts);

  // Call the partitioning function
  this->_do_repartition(mesh,n_parts);

  // Set the parent's processor ids
  Partitioner::set_parent_processor_ids(mesh);

  // Set the node's processor ids
  Partitioner::set_node_processor_ids(mesh);
}

repartition() [2/2]

void libMesh::Partitioner::repartition ( MeshBase &  mesh )

inherited

Repartitions the MeshBase into mesh.n_processors() parts. This is required since some partitioning algorithms can repartition more efficiently than computing a new partitioning from scratch.

Definition at line 117 of file partitioner.C.

References mesh, and libMesh::Partitioner::repartition().

{
  this->repartition(mesh,mesh.n_processors());
}

set_interface_node_processor_ids_BFS()

void libMesh::Partitioner::set_interface_node_processor_ids_BFS ( MeshBase &  mesh )

staticinherited

Nodes on the partitioning interface is clustered into two groups BFS (Breadth First Search)scheme for per pair of processors

Definition at line 498 of file partitioner.C.

References libMesh::MeshTools::build_nodes_to_elem_map(), libMesh::MeshTools::find_nodal_neighbors(), mesh, and libMesh::Partitioner::processor_pairs_to_interface_nodes().

Referenced by libMesh::Partitioner::set_node_processor_ids().

{
  // This function must be run on all processors at once
  libmesh_parallel_only(mesh.comm());

  std::map<std::pair<processor_id_type, processor_id_type>, std::set<dof_id_type>> processor_pair_to_nodes;

  processor_pairs_to_interface_nodes(mesh, processor_pair_to_nodes);

  std::unordered_map<dof_id_type, std::vector<const Elem *>> nodes_to_elem_map;

  MeshTools::build_nodes_to_elem_map(mesh, nodes_to_elem_map);

  std::vector<const Node *>  neighbors;
  std::set<dof_id_type> neighbors_order;
  std::vector<dof_id_type> common_nodes;
  std::queue<dof_id_type> nodes_queue;
  std::set<dof_id_type> visted_nodes;

  for (auto & pmap : processor_pair_to_nodes)
    {
      std::size_t n_own_nodes = pmap.second.size()/2;

      // Initialize node assignment
      for (auto it = pmap.second.begin(); it != pmap.second.end(); it++)
        mesh.node_ref(*it).processor_id() = pmap.first.second;

      visted_nodes.clear();
      for (auto it = pmap.second.begin(); it != pmap.second.end(); it++)
        {
          mesh.node_ref(*it).processor_id() = pmap.first.second;

          if (visted_nodes.find(*it) != visted_nodes.end())
            continue;
          else
            {
              nodes_queue.push(*it);
              visted_nodes.insert(*it);
              if (visted_nodes.size() >= n_own_nodes)
                break;
            }

          while (!nodes_queue.empty())
            {
              auto & node = mesh.node_ref(nodes_queue.front());
              nodes_queue.pop();

              neighbors.clear();
              MeshTools::find_nodal_neighbors(mesh, node, nodes_to_elem_map, neighbors);
              neighbors_order.clear();
              for (auto & neighbor : neighbors)
                neighbors_order.insert(neighbor->id());

              common_nodes.clear();
              std::set_intersection(pmap.second.begin(), pmap.second.end(),
                                    neighbors_order.begin(), neighbors_order.end(),
                                    std::back_inserter(common_nodes));

              for (auto c_node : common_nodes)
                if (visted_nodes.find(c_node) == visted_nodes.end())
                  {
                    nodes_queue.push(c_node);
                    visted_nodes.insert(c_node);
                    if (visted_nodes.size() >= n_own_nodes)
                      goto queue_done;
                  }

              if (visted_nodes.size() >= n_own_nodes)
                goto queue_done;
            }
        }
    queue_done:
      for (auto node : visted_nodes)
        mesh.node_ref(node).processor_id() = pmap.first.first;
    }
}

set_interface_node_processor_ids_linear()

void libMesh::Partitioner::set_interface_node_processor_ids_linear ( MeshBase &  mesh )

staticinherited

Nodes on the partitioning interface is linearly assigned to each pair of processors

Definition at line 474 of file partitioner.C.

References mesh, and libMesh::Partitioner::processor_pairs_to_interface_nodes().

Referenced by libMesh::Partitioner::set_node_processor_ids().

{
  // This function must be run on all processors at once
  libmesh_parallel_only(mesh.comm());

  std::map<std::pair<processor_id_type, processor_id_type>, std::set<dof_id_type>> processor_pair_to_nodes;

  processor_pairs_to_interface_nodes(mesh, processor_pair_to_nodes);

  for (auto & pmap : processor_pair_to_nodes)
    {
      std::size_t n_own_nodes = pmap.second.size()/2, i = 0;

      for (auto it = pmap.second.begin(); it != pmap.second.end(); it++, i++)
        {
          auto & node = mesh.node_ref(*it);
          if (i <= n_own_nodes)
            node.processor_id() = pmap.first.first;
          else
            node.processor_id() = pmap.first.second;
        }
    }
}

set_interface_node_processor_ids_petscpartitioner()

void libMesh::Partitioner::set_interface_node_processor_ids_petscpartitioner ( MeshBase &  mesh )

staticinherited

Nodes on the partitioning interface is partitioned into two groups using a PETSc partitioner for each pair of processors

Definition at line 575 of file partitioner.C.

References libMesh::MeshTools::build_nodes_to_elem_map(), libMesh::MeshTools::find_nodal_neighbors(), libMesh::libmesh_ignore(), mesh, and libMesh::Partitioner::processor_pairs_to_interface_nodes().

Referenced by libMesh::Partitioner::set_node_processor_ids().

{
  libmesh_ignore(mesh); // Only used if LIBMESH_HAVE_PETSC

  // This function must be run on all processors at once
  libmesh_parallel_only(mesh.comm());

#if LIBMESH_HAVE_PETSC
  std::map<std::pair<processor_id_type, processor_id_type>, std::set<dof_id_type>> processor_pair_to_nodes;

  processor_pairs_to_interface_nodes(mesh, processor_pair_to_nodes);

  std::vector<std::vector<const Elem *>> nodes_to_elem_map;

  MeshTools::build_nodes_to_elem_map(mesh, nodes_to_elem_map);

  std::vector<const Node *>  neighbors;
  std::set<dof_id_type> neighbors_order;
  std::vector<dof_id_type> common_nodes;

  std::vector<dof_id_type> rows;
  std::vector<dof_id_type> cols;

  std::map<dof_id_type, dof_id_type> global_to_local;

  for (auto & pmap : processor_pair_to_nodes)
    {
      unsigned int i = 0;

      rows.clear();
      rows.resize(pmap.second.size()+1);
      cols.clear();
      for (auto it = pmap.second.begin(); it != pmap.second.end(); it++)
        global_to_local[*it] = i++;

      i = 0;
      for (auto it = pmap.second.begin(); it != pmap.second.end(); it++, i++)
        {
          auto & node = mesh.node_ref(*it);
          neighbors.clear();
          MeshTools::find_nodal_neighbors(mesh, node, nodes_to_elem_map, neighbors);
          neighbors_order.clear();
          for (auto & neighbor : neighbors)
            neighbors_order.insert(neighbor->id());

          common_nodes.clear();
          std::set_intersection(pmap.second.begin(), pmap.second.end(),
                                neighbors_order.begin(), neighbors_order.end(),
                                std::back_inserter(common_nodes));

          rows[i+1] = rows[i] + cast_int<dof_id_type>(common_nodes.size());

          for (auto c_node : common_nodes)
            cols.push_back(global_to_local[c_node]);
        }

      Mat adj;
      MatPartitioning part;
      IS is;
      PetscInt local_size, rows_size, cols_size;
      PetscInt *adj_i, *adj_j;
      const PetscInt *indices;
      PetscCalloc1(rows.size(), &adj_i);
      PetscCalloc1(cols.size(), &adj_j);
      rows_size = cast_int<PetscInt>(rows.size());
      for (PetscInt ii=0; ii<rows_size; ii++)
        adj_i[ii] = rows[ii];

      cols_size = cast_int<PetscInt>(cols.size());
      for (PetscInt ii=0; ii<cols_size; ii++)
        adj_j[ii] = cols[ii];

      const PetscInt sz = cast_int<PetscInt>(pmap.second.size());
      MatCreateMPIAdj(PETSC_COMM_SELF, sz, sz, adj_i, adj_j,nullptr,&adj);
      MatPartitioningCreate(PETSC_COMM_SELF,&part);
      MatPartitioningSetAdjacency(part,adj);
      MatPartitioningSetNParts(part,2);
      PetscObjectSetOptionsPrefix((PetscObject)part, "balance_");
      MatPartitioningSetFromOptions(part);
      MatPartitioningApply(part,&is);

      MatDestroy(&adj);
      MatPartitioningDestroy(&part);

      ISGetLocalSize(is, &local_size);
      ISGetIndices(is, &indices);
      i = 0;
      for (auto it = pmap.second.begin(); it != pmap.second.end(); it++, i++)
        {
          auto & node = mesh.node_ref(*it);
          if (indices[i])
            node.processor_id() = pmap.first.second;
          else
            node.processor_id() = pmap.first.first;
        }
      ISRestoreIndices(is, &indices);
      ISDestroy(&is);
    }
#else
  libmesh_error_msg("PETSc is required");
#endif
}

set_node_processor_ids()

void libMesh::Partitioner::set_node_processor_ids ( MeshBase &  mesh )

staticinherited

This function is called after partitioning to set the processor IDs for the nodes. By definition, a Node's processor ID is the minimum processor ID for all of the elements which share the node.

Definition at line 679 of file partitioner.C.

References libMesh::as_range(), libMesh::Node::choose_processor_id(), libMesh::DofObject::invalid_processor_id, mesh, libMesh::MeshTools::n_elem(), libMesh::on_command_line(), libMesh::DofObject::processor_id(), libMesh::Parallel::pull_parallel_vector_data(), libMesh::Partitioner::set_interface_node_processor_ids_BFS(), libMesh::Partitioner::set_interface_node_processor_ids_linear(), and libMesh::Partitioner::set_interface_node_processor_ids_petscpartitioner().

Referenced by libMesh::MeshRefinement::_refine_elements(), libMesh::UnstructuredMesh::all_first_order(), libMesh::Partitioner::partition(), libMesh::XdrIO::read(), libMesh::Partitioner::repartition(), and libMesh::BoundaryInfo::sync().

{
  LOG_SCOPE("set_node_processor_ids()","Partitioner");

  // This function must be run on all processors at once
  libmesh_parallel_only(mesh.comm());

  // If we have any unpartitioned elements at this
  // stage there is a problem
  libmesh_assert (MeshTools::n_elem(mesh.unpartitioned_elements_begin(),
                                    mesh.unpartitioned_elements_end()) == 0);


  //   const dof_id_type orig_n_local_nodes = mesh.n_local_nodes();

  //   libMesh::err << "[" << mesh.processor_id() << "]: orig_n_local_nodes="
  //     << orig_n_local_nodes << std::endl;

  // Build up request sets.  Each node is currently owned by a processor because
  // it is connected to an element owned by that processor.  However, during the
  // repartitioning phase that element may have been assigned a new processor id, but
  // it is still resident on the original processor.  We need to know where to look
  // for new ids before assigning new ids, otherwise we may be asking the wrong processors
  // for the wrong information.
  //
  // The only remaining issue is what to do with unpartitioned nodes.  Since they are required
  // to live on all processors we can simply rely on ourselves to number them properly.
  std::map<processor_id_type, std::vector<dof_id_type>>
    requested_node_ids;

  // Loop over all the nodes, count the ones on each processor.  We can skip ourself
  std::vector<dof_id_type> ghost_nodes_from_proc(mesh.n_processors(), 0);

  for (auto & node : mesh.node_ptr_range())
    {
      libmesh_assert(node);
      const processor_id_type current_pid = node->processor_id();
      if (current_pid != mesh.processor_id() &&
          current_pid != DofObject::invalid_processor_id)
        {
          libmesh_assert_less (current_pid, ghost_nodes_from_proc.size());
          ghost_nodes_from_proc[current_pid]++;
        }
    }

  // We know how many objects live on each processor, so reserve()
  // space for each.
  for (processor_id_type pid=0; pid != mesh.n_processors(); ++pid)
    if (ghost_nodes_from_proc[pid])
      requested_node_ids[pid].reserve(ghost_nodes_from_proc[pid]);

  // We need to get the new pid for each node from the processor
  // which *currently* owns the node.  We can safely skip ourself
  for (auto & node : mesh.node_ptr_range())
    {
      libmesh_assert(node);
      const processor_id_type current_pid = node->processor_id();
      if (current_pid != mesh.processor_id() &&
          current_pid != DofObject::invalid_processor_id)
        {
          libmesh_assert_less (requested_node_ids[current_pid].size(),
                               ghost_nodes_from_proc[current_pid]);
          requested_node_ids[current_pid].push_back(node->id());
        }

      // Unset any previously-set node processor ids
      node->invalidate_processor_id();
    }

  // Loop over all the active elements
  for (auto & elem : mesh.active_element_ptr_range())
    {
      libmesh_assert(elem);

      libmesh_assert_not_equal_to (elem->processor_id(), DofObject::invalid_processor_id);

      // Consider updating the processor id on this element's nodes
      for (unsigned int n=0; n<elem->n_nodes(); ++n)
        {
          Node & node = elem->node_ref(n);
          processor_id_type & pid = node.processor_id();
          pid = node.choose_processor_id(pid, elem->processor_id());
        }
    }

  bool load_balanced_nodes_linear =
      libMesh::on_command_line ("--load-balanced-nodes-linear");

  if (load_balanced_nodes_linear)
    set_interface_node_processor_ids_linear(mesh);

  bool load_balanced_nodes_bfs =
       libMesh::on_command_line ("--load-balanced-nodes-bfs");

  if (load_balanced_nodes_bfs)
    set_interface_node_processor_ids_BFS(mesh);

  bool load_balanced_nodes_petscpartition =
      libMesh::on_command_line ("--load_balanced_nodes_petscpartitioner");

  if (load_balanced_nodes_petscpartition)
    set_interface_node_processor_ids_petscpartitioner(mesh);

  // And loop over the subactive elements, but don't reassign
  // nodes that are already active on another processor.
  for (auto & elem : as_range(mesh.subactive_elements_begin(),
                              mesh.subactive_elements_end()))
    {
      libmesh_assert(elem);

      libmesh_assert_not_equal_to (elem->processor_id(), DofObject::invalid_processor_id);

      for (unsigned int n=0; n<elem->n_nodes(); ++n)
        if (elem->node_ptr(n)->processor_id() == DofObject::invalid_processor_id)
          elem->node_ptr(n)->processor_id() = elem->processor_id();
    }

  // Same for the inactive elements -- we will have already gotten most of these
  // nodes, *except* for the case of a parent with a subset of children which are
  // ghost elements.  In that case some of the parent nodes will not have been
  // properly handled yet
  for (auto & elem : as_range(mesh.not_active_elements_begin(),
                              mesh.not_active_elements_end()))
    {
      libmesh_assert(elem);

      libmesh_assert_not_equal_to (elem->processor_id(), DofObject::invalid_processor_id);

      for (unsigned int n=0; n<elem->n_nodes(); ++n)
        if (elem->node_ptr(n)->processor_id() == DofObject::invalid_processor_id)
          elem->node_ptr(n)->processor_id() = elem->processor_id();
    }

  // We can't assert that all nodes are connected to elements, because
  // a DistributedMesh with NodeConstraints might have pulled in some
  // remote nodes solely for evaluating those constraints.
  // MeshTools::libmesh_assert_connected_nodes(mesh);

  // For such nodes, we'll do a sanity check later when making sure
  // that we successfully reset their processor ids to something
  // valid.

  auto gather_functor =
    [& mesh]
    (processor_id_type, const std::vector<dof_id_type> & ids,
     std::vector<processor_id_type> & new_pids)
    {
      const std::size_t ids_size = ids.size();
      new_pids.resize(ids_size);

      // Fill those requests in-place
      for (std::size_t i=0; i != ids_size; ++i)
        {
          Node & node = mesh.node_ref(ids[i]);
          const processor_id_type new_pid = node.processor_id();

          // We may have an invalid processor_id() on nodes that have been
          // "detached" from coarsened-away elements but that have not yet
          // themselves been removed.
          // libmesh_assert_not_equal_to (new_pid, DofObject::invalid_processor_id);
          // libmesh_assert_less (new_pid, mesh.n_partitions()); // this is the correct test --
          new_pids[i] = new_pid;                                 //  the number of partitions may
        }                                                        //  not equal the number of processors
    };

  auto action_functor =
    [& mesh]
    (processor_id_type,
     const std::vector<dof_id_type> & ids,
     const std::vector<processor_id_type> & new_pids)
    {
      const std::size_t ids_size = ids.size();
      // Copy the pid changes we've now been informed of
      for (std::size_t i=0; i != ids_size; ++i)
        {
          Node & node = mesh.node_ref(ids[i]);

          // this is the correct test -- the number of partitions may
          // not equal the number of processors

          // But: we may have an invalid processor_id() on nodes that
          // have been "detached" from coarsened-away elements but
          // that have not yet themselves been removed.
          // libmesh_assert_less (filled_request[i], mesh.n_partitions());

          node.processor_id(new_pids[i]);
        }
    };

  const processor_id_type * ex = nullptr;
  Parallel::pull_parallel_vector_data
    (mesh.comm(), requested_node_ids, gather_functor, action_functor, ex);

#ifdef DEBUG
  MeshTools::libmesh_assert_valid_procids<Node>(mesh);
  //MeshTools::libmesh_assert_canonical_node_procids(mesh);
#endif
}

set_parent_processor_ids()

void libMesh::Partitioner::set_parent_processor_ids ( MeshBase &  mesh )

staticinherited

This function is called after partitioning to set the processor IDs for the inactive parent elements. A parent's processor ID is the same as its first child.

Definition at line 268 of file partitioner.C.

References libMesh::as_range(), libMesh::Elem::child_ref_range(), libMesh::Partitioner::communication_blocksize, libMesh::DofObject::invalid_processor_id, libMesh::DofObject::invalidate_processor_id(), libMesh::libmesh_ignore(), mesh, std::min(), libMesh::MeshTools::n_elem(), libMesh::Elem::parent(), libMesh::DofObject::processor_id(), and libMesh::Elem::total_family_tree().

Referenced by libMesh::Partitioner::partition(), and libMesh::Partitioner::repartition().

{
  // Ignore the parameter when !LIBMESH_ENABLE_AMR
  libmesh_ignore(mesh);

  LOG_SCOPE("set_parent_processor_ids()", "Partitioner");

#ifdef LIBMESH_ENABLE_AMR

  // If the mesh is serial we have access to all the elements,
  // in particular all the active ones.  We can therefore set
  // the parent processor ids indirectly through their children, and
  // set the subactive processor ids while examining their active
  // ancestors.
  // By convention a parent is assigned to the minimum processor
  // of all its children, and a subactive is assigned to the processor
  // of its active ancestor.
  if (mesh.is_serial())
    {
      for (auto & elem : mesh.active_element_ptr_range())
        {
          // First set descendents
          std::vector<const Elem *> subactive_family;
          elem->total_family_tree(subactive_family);
          for (std::size_t i = 0; i != subactive_family.size(); ++i)
            const_cast<Elem *>(subactive_family[i])->processor_id() = elem->processor_id();

          // Then set ancestors
          Elem * parent = elem->parent();

          while (parent)
            {
              // invalidate the parent id, otherwise the min below
              // will not work if the current parent id is less
              // than all the children!
              parent->invalidate_processor_id();

              for (auto & child : parent->child_ref_range())
                {
                  libmesh_assert(!child.is_remote());
                  libmesh_assert_not_equal_to (child.processor_id(), DofObject::invalid_processor_id);
                  parent->processor_id() = std::min(parent->processor_id(),
                                                    child.processor_id());
                }
              parent = parent->parent();
            }
        }
    }

  // When the mesh is parallel we cannot guarantee that parents have access to
  // all their children.
  else
    {
      // Setting subactive processor ids is easy: we can guarantee
      // that children have access to all their parents.

      // Loop over all the active elements in the mesh
      for (auto & child : mesh.active_element_ptr_range())
        {
          std::vector<const Elem *> subactive_family;
          child->total_family_tree(subactive_family);
          for (std::size_t i = 0; i != subactive_family.size(); ++i)
            const_cast<Elem *>(subactive_family[i])->processor_id() = child->processor_id();
        }

      // When the mesh is parallel we cannot guarantee that parents have access to
      // all their children.

      // We will use a brute-force approach here.  Each processor finds its parent
      // elements and sets the parent pid to the minimum of its
      // semilocal descendants.
      // A global reduction is then performed to make sure the true minimum is found.
      // As noted, this is required because we cannot guarantee that a parent has
      // access to all its children on any single processor.
      libmesh_parallel_only(mesh.comm());
      libmesh_assert(MeshTools::n_elem(mesh.unpartitioned_elements_begin(),
                                       mesh.unpartitioned_elements_end()) == 0);

      const dof_id_type max_elem_id = mesh.max_elem_id();

      std::vector<processor_id_type>
        parent_processor_ids (std::min(communication_blocksize,
                                       max_elem_id));

      for (dof_id_type blk=0, last_elem_id=0; last_elem_id<max_elem_id; blk++)
        {
          last_elem_id =
            std::min(static_cast<dof_id_type>((blk+1)*communication_blocksize),
                     max_elem_id);
          const dof_id_type first_elem_id = blk*communication_blocksize;

          std::fill (parent_processor_ids.begin(),
                     parent_processor_ids.end(),
                     DofObject::invalid_processor_id);

          // first build up local contributions to parent_processor_ids
          bool have_parent_in_block = false;

          for (auto & parent : as_range(mesh.ancestor_elements_begin(),
                                        mesh.ancestor_elements_end()))
            {
              const dof_id_type parent_idx = parent->id();
              libmesh_assert_less (parent_idx, max_elem_id);

              if ((parent_idx >= first_elem_id) &&
                  (parent_idx <  last_elem_id))
                {
                  have_parent_in_block = true;
                  processor_id_type parent_pid = DofObject::invalid_processor_id;

                  std::vector<const Elem *> active_family;
                  parent->active_family_tree(active_family);
                  for (std::size_t i = 0; i != active_family.size(); ++i)
                    parent_pid = std::min (parent_pid, active_family[i]->processor_id());

                  const dof_id_type packed_idx = parent_idx - first_elem_id;
                  libmesh_assert_less (packed_idx, parent_processor_ids.size());

                  parent_processor_ids[packed_idx] = parent_pid;
                }
            }

          // then find the global minimum
          mesh.comm().min (parent_processor_ids);

          // and assign the ids, if we have a parent in this block.
          if (have_parent_in_block)
            for (auto & parent : as_range(mesh.ancestor_elements_begin(),
                                          mesh.ancestor_elements_end()))
              {
                const dof_id_type parent_idx = parent->id();

                if ((parent_idx >= first_elem_id) &&
                    (parent_idx <  last_elem_id))
                  {
                    const dof_id_type packed_idx = parent_idx - first_elem_id;
                    libmesh_assert_less (packed_idx, parent_processor_ids.size());

                    const processor_id_type parent_pid =
                      parent_processor_ids[packed_idx];

                    libmesh_assert_not_equal_to (parent_pid, DofObject::invalid_processor_id);

                    parent->processor_id() = parent_pid;
                  }
              }
        }
    }

#endif // LIBMESH_ENABLE_AMR
}

single_partition()

void libMesh::Partitioner::single_partition ( MeshBase &  mesh )

protectedinherited

Trivially "partitions" the mesh for one processor. Simply loops through the elements and assigns all of them to processor 0. Is is provided as a separate function so that derived classes may use it without reimplementing it.

Definition at line 159 of file partitioner.C.

References libMesh::MeshBase::elements_begin(), mesh, and libMesh::Partitioner::single_partition_range().

Referenced by _do_partition(), libMesh::Partitioner::partition(), and libMesh::Partitioner::repartition().

{
  this->single_partition_range(mesh.elements_begin(),
                               mesh.elements_end());

  // Redistribute, in case someone (like our unit tests) is doing
  // something silly (like moving a whole already-distributed mesh
  // back onto rank 0).
  mesh.redistribute();
}

single_partition_range()

void libMesh::Partitioner::single_partition_range ( MeshBase::element_iterator  it, MeshBase::element_iterator  end  )

protectedinherited

Slightly generalized version of single_partition which acts on a range of elements defined by the pair of iterators (it, end).

Definition at line 172 of file partitioner.C.

References libMesh::as_range(), and end.

Referenced by libMesh::LinearPartitioner::partition_range(), libMesh::MetisPartitioner::partition_range(), libMesh::MappedSubdomainPartitioner::partition_range(), libMesh::SFCPartitioner::partition_range(), libMesh::CentroidPartitioner::partition_range(), and libMesh::Partitioner::single_partition().

{
  LOG_SCOPE("single_partition_range()", "Partitioner");

  for (auto & elem : as_range(it, end))
    {
      elem->processor_id() = 0;

      // Assign all this element's nodes to processor 0 as well.
      for (unsigned int n=0; n<elem->n_nodes(); ++n)
        elem->node_ptr(n)->processor_id() = 0;
    }
}

Member Data Documentation

_dual_graph

std::vector<std::vector<dof_id_type> > libMesh::Partitioner::_dual_graph

protectedinherited

A dual graph corresponds to the mesh, and it is typically used in paritioner. A vertex represents an element, and its neighbors are the element neighbors.

Definition at line 288 of file partitioner.h.

Referenced by libMesh::Partitioner::build_graph().

_global_index_by_pid_map

std::unordered_map<dof_id_type, dof_id_type> libMesh::Partitioner::_global_index_by_pid_map

protectedinherited

Maps active element ids into a contiguous range, as needed by parallel partitioner.

Definition at line 272 of file partitioner.h.

Referenced by libMesh::Partitioner::_find_global_index_by_pid_map(), libMesh::Partitioner::assign_partitioning(), and libMesh::Partitioner::build_graph().

_internal_partitioner

std::unique_ptr<Partitioner> libMesh::SubdomainPartitioner::_internal_partitioner

protected

The internal Partitioner we use. Public access via the internal_partitioner() member function.

Definition at line 118 of file subdomain_partitioner.h.

Referenced by _do_partition(), and internal_partitioner().

_local_id_to_elem

std::vector<Elem *> libMesh::Partitioner::_local_id_to_elem

protectedinherited

Definition at line 291 of file partitioner.h.

Referenced by libMesh::Partitioner::build_graph().

_n_active_elem_on_proc

std::vector<dof_id_type> libMesh::Partitioner::_n_active_elem_on_proc

protectedinherited

The number of active elements on each processor.

Note

ParMETIS requires that each processor have some active elements; it will abort if any processor passes a nullptr _part array.

Definition at line 281 of file partitioner.h.

Referenced by libMesh::Partitioner::_find_global_index_by_pid_map(), libMesh::Partitioner::assign_partitioning(), and libMesh::Partitioner::build_graph().

_weights

ErrorVector* libMesh::Partitioner::_weights

protectedinherited

The weights that might be used for partitioning.

Definition at line 267 of file partitioner.h.

Referenced by libMesh::MetisPartitioner::attach_weights(), and libMesh::MetisPartitioner::partition_range().

chunks

std::vector<std::set<subdomain_id_type> > libMesh::SubdomainPartitioner::chunks

Each entry of "chunks" represents a set of subdomains which are to be partitioned together. The internal Partitioner will be called once for each entry of chunks, and the resulting partitioning will simply be the union of all these partitionings.

Definition at line 95 of file subdomain_partitioner.h.

Referenced by _do_partition().

communication_blocksize

const dof_id_type libMesh::Partitioner::communication_blocksize = 1000000

staticprotectedinherited

The blocksize to use when doing blocked parallel communication. This limits the maximum vector size which can be used in a single communication step.

Definition at line 244 of file partitioner.h.

Referenced by libMesh::Partitioner::set_parent_processor_ids().

---

The documentation for this class was generated from the following files:

• subdomain_partitioner.h
• subdomain_partitioner.C