ArborX / Spatial indexes / ArborX::DistributedTree

`ArborX::DistributedTree<MemorySpace, Value, IndexableGetter, BoundingVolume>::DistributedTree`¶

DistributedTree() noexcept; // (1)

template <typename ExecutionSpace, typename Values>
DistributedTree(MPI_Comm comm, ExecutionSpace const& space, Values const& values); // (2)

template <typename ExecutionSpace, typename Values>
DistributedTree(MPI_Comm comm, ExecutionSpace const& space, Values const& values, IndexableGetter const& indexable_getter); // (3)

Default constructor. Constructs an empty tree.
Constructs a bounding volume hierarchy from a given data source.
Constructs a bounding volume hierarchy from a given data source and a function tranforming a value into indexable.

Parameters¶

comm : the MPI communicator.

space: execution space values: values stored in the container indexable_getter: a function object to transform a value into an indexable

Type requirements¶

MemorySpace must be accessible from ExecutionSpace (i.e., Kokkos::SpaceAccessibility<ExecutionSpace, MemorySpace>::accessible must be true).
A specialization of ArborX::AccessTraits must match Values as the template argument.
The return type of ArborX::AccessTraits<Values>::get() must decay to Value.

ArborX provides specializations for Kokkos views but a user may specialize it for their types.
IndexableGetter must provide an operator() taking in a type that decays to Value and returning an indexable (geometric object).

Complexity¶

O(N log N), where N is the number of primitives passed to the constructor (ArborX::AccessTraits<Values>::size(values)).

Exceptions¶

Memory allocation with Kokkos may throw.

Notes¶

The constructor must be called by all MPI ranks related in the provided MPI communicator.

Example¶

#include <ArborX_DistributedTree.hpp>

#include <Kokkos_Core.hpp>

#include <iostream>

int main(int argc, char *argv[])
{
  MPI_Init(&argc, &argv);
  Kokkos::initialize(argc, argv);

  MPI_Comm const comm = MPI_COMM_WORLD;
  int comm_rank;
  MPI_Comm_rank(comm, &comm_rank);

  Kokkos::View<ArborX::Point<3> *> cloud("point_cloud", 10);
  Kokkos::parallel_for(10, KOKKOS_LAMBDA(int i) {
    cloud[i] = {{(float)i + 10 * comm_rank, (float)i, (float)i}};
  });

  using memory_space = decltype(cloud)::memory_space; // where to store the tree
  using execution_space = decltype(cloud)::execution_space; // where to execute code
  ArborX::DistributedTree dist_tree{comm, execution_space{}, cloud};

  if (comm_rank == 0)
  {
    auto const box = dist_tree.bounds();
    auto min = box.minCorner();
    auto max = box.maxCorner();
    std::cout << min[0] << ',' << min[1] << ',' << min[2] << " - "
              << max[0] << ',' << max[1] << ',' << max[2] << '\n';
  }

  Kokkos::finalize();
  MPI_Finalize();

  return 0;
}

Output for 4 MPI processes

(0,0,0) - (39,9,9)

ArborX::DistributedTree<MemorySpace, Value, IndexableGetter, BoundingVolume>::DistributedTree¶