Template Numerical Library version\ main:c173bea
Loading...
Searching...
No Matches
TNL::Containers::DistributedNDArray< NDArray > Class Template Reference

Distributed N-dimensional array. More...

#include <TNL/Containers/DistributedNDArray.h>

Collaboration diagram for TNL::Containers::DistributedNDArray< NDArray >:

Public Types

using AllocatorType = typename NDArray::AllocatorType
 Allocator type used for allocating the array.
 
using ConstLocalViewType = typename NDArray::ConstViewType
 Compatible constant NDArrayView of the local array.
 
using ConstViewType = DistributedNDArrayView< typename NDArray::ConstViewType >
 Compatible constant DistributedNDArrayView type.
 
using DeviceType = typename NDArray::DeviceType
 Type of the device used for running operations on the array.
 
using IndexType = typename NDArray::IndexType
 Type of indices used for addressing the array elements.
 
using LocalBeginsType = LocalBeginsHolder< typename NDArray::SizesHolderType >
 Type which represents the position of the first local element in the global N-dimensional array. It has all static sizes set to 0.
 
using LocalRangeType = Subrange< IndexType >
 Type which represents an integer range [a, b).
 
using LocalViewType = typename NDArray::ViewType
 Compatible NDArrayView of the local array.
 
using OverlapsType = typename NDArray::OverlapsType
 Sequence of integers representing the overlaps in each dimension of a distributed N-dimensional array.
 
using PermutationType = typename NDArray::PermutationType
 Permutation that is applied to indices when accessing the array elements.
 
using SizesHolderType = typename NDArray::SizesHolderType
 Type of the underlying object which represents the sizes of the N-dimensional array.
 
using ValueType = typename NDArray::ValueType
 Type of the values stored in the array.
 
using ViewType = DistributedNDArrayView< typename NDArray::ViewType >
 Compatible DistributedNDArrayView type.
 

Public Member Functions

 DistributedNDArray ()=default
 Constructs an empty array with zero size.
 
 DistributedNDArray (const AllocatorType &allocator)
 Constructs an empty array and sets the provided allocator.
 
 DistributedNDArray (const DistributedNDArray &)=default
 Copy constructor (makes a deep copy).
 
 DistributedNDArray (const DistributedNDArray &other, const AllocatorType &allocator)
 Copy constructor with a specific allocator (makes a deep copy).
 
 DistributedNDArray (DistributedNDArray &&) noexcept=default
 Move constructor for initialization from rvalues.
 
void allocate ()
 Computes the distributed storage size and allocates the local array.
 
template<typename Device2 = DeviceType, typename Func >
void forAll (Func f, const typename Device2::LaunchConfiguration &launch_configuration=typename Device2::LaunchConfiguration{}) const
 Evaluates the function f in parallel for all elements of the array.
 
template<typename Device2 = DeviceType, typename SkipBegins , typename SkipEnds , typename Func >
void forBoundary (const SkipBegins &skipBegins, const SkipEnds &skipEnds, Func f, const typename Device2::LaunchConfiguration &launch_configuration=typename Device2::LaunchConfiguration{}) const
 Evaluates the function f in parallel for all elements outside the given [skipBegins, skipEnds) range specified by global indices.
 
template<typename Device2 = DeviceType, typename Func >
void forBoundary (Func f, const typename Device2::LaunchConfiguration &launch_configuration=typename Device2::LaunchConfiguration{}) const
 Evaluates the function f in parallel for all boundary elements of the array.
 
template<typename Device2 = DeviceType, typename Func >
void forGhosts (Func f, const typename Device2::LaunchConfiguration &launch_configuration=typename Device2::LaunchConfiguration{}) const
 Evaluates the function f in parallel for all elements in the ghost region.
 
template<typename Device2 = DeviceType, typename Begins , typename Ends , typename Func >
void forInterior (const Begins &begins, const Ends &ends, Func f, const typename Device2::LaunchConfiguration &launch_configuration=typename Device2::LaunchConfiguration{}) const
 Evaluates the function f in parallel for all elements inside the given [begins, ends) range specified by global indices.
 
template<typename Device2 = DeviceType, typename Func >
void forInterior (Func f, const typename Device2::LaunchConfiguration &launch_configuration=typename Device2::LaunchConfiguration{}) const
 Evaluates the function f in parallel for all internal elements of the array.
 
template<typename Device2 = DeviceType, typename Func >
void forLocalBoundary (Func f, const typename Device2::LaunchConfiguration &launch_configuration=typename Device2::LaunchConfiguration{}) const
 Evaluates the function f in parallel for all local-boundary elements of the array.
 
template<typename Device2 = DeviceType, typename Func >
void forLocalInterior (Func f, const typename Device2::LaunchConfiguration &launch_configuration=typename Device2::LaunchConfiguration{}) const
 Evaluates the function f in parallel for all local-internal elements of the array.
 
AllocatorType getAllocator () const
 Returns the allocator associated with the array.
 
const MPI::CommgetCommunicator () const
 Returns the MPI communicator associated with the array.
 
ConstLocalViewType getConstLocalView () const
 Returns a non-modifiable view of the local array.
 
ConstViewType getConstView () const
 Returns a non-modifiable view of the array.
 
ValueTypegetData ()
 Returns a raw pointer to the local data.
 
std::add_const_t< ValueType > * getData () const
 Returns a const-qualified raw pointer to the local data.
 
template<typename... IndexTypes>
ValueType getElement (IndexTypes &&... indices) const
 A "safe" accessor for array elements.
 
LocalBeginsType getLocalBegins () const
 Returns the beginning position of the local array in the global N-dimensional array.
 
SizesHolderType getLocalEnds () const
 Returns the ending position of the local array in the global N-dimensional array.
 
template<std::size_t level>
LocalRangeType getLocalRange () const
 Returns a specific [begin, end) subrange of the local array in the global N-dimensional array.
 
IndexType getLocalStorageSize () const
 Returns the size (number of elements) needed to store the local N-dimensional array.
 
LocalViewType getLocalView ()
 Returns a modifiable view of the local array.
 
template<std::size_t level>
__cuda_callable__ IndexType getOverlap () const
 Returns the overlap of a distributed N-dimensional array along the specified axis.
 
__cuda_callable__ OverlapsTypegetOverlaps ()
 Returns the N-dimensional overlaps holder instance.
 
__cuda_callable__ const OverlapsTypegetOverlaps () const
 Returns the N-dimensional overlaps holder instance.
 
template<std::size_t level>
IndexType getSize () const
 Returns a specific component of the N-dimensional global sizes.
 
const SizesHolderTypegetSizes () const
 Returns the N-dimensional sizes of the global array.
 
template<typename... IndexTypes>
IndexType getStorageIndex (IndexTypes &&... indices) const
 Returns the local storage index for given global indices.
 
ViewType getView ()
 Returns a modifiable view of the array.
 
bool operator!= (const DistributedNDArray &other) const
 Compares the array with another distributed N-dimensional array.
 
template<typename... IndexTypes>
ValueTypeoperator() (IndexTypes &&... indices)
 Accesses an element of the array.
 
template<typename... IndexTypes>
const ValueTypeoperator() (IndexTypes &&... indices) const
 Accesses an element of the array.
 
DistributedNDArrayoperator= (const DistributedNDArray &other)=default
 Copy-assignment operator for deep-copying data from another array. Mismatched sizes cause reallocations.
 
template<typename OtherArray >
DistributedNDArrayoperator= (const OtherArray &other)
 Templated copy-assignment operator for deep-copying data from another array.
 
DistributedNDArrayoperator= (DistributedNDArray &&) noexcept(false)=default
 Move-assignment operator for acquiring data from rvalues.
 
bool operator== (const DistributedNDArray &other) const
 Compares the array with another distributed N-dimensional array.
 
ValueTypeoperator[] (IndexType index)
 Accesses an element in a one-dimensional array.
 
const ValueTypeoperator[] (IndexType index) const
 Accesses an element in a one-dimensional array.
 
void reset ()
 Resets the array to the empty state.
 
template<std::size_t level>
void setDistribution (IndexType begin, IndexType end, const MPI::Comm &communicator=MPI_COMM_WORLD)
 Sets the distribution of the arrray among MPI ranks along the given axis.
 
void setLike (const DistributedNDArray &other)
 Sets sizes of the array to the sizes of an existing array.
 
template<typename... IndexTypes>
void setSizes (IndexTypes &&... sizes)
 Sets the global sizes of the array, but does not allocate storage.
 
void setValue (ValueType value)
 Sets all elements of the array to given value.
 

Static Public Member Functions

static constexpr std::size_t getDimension ()
 Returns the dimension of the N-dimensional array, i.e. N.
 

Protected Attributes

MPI::Comm communicator = MPI_COMM_NULL
 MPI communicator associated with the array.
 
SizesHolderType globalSizes
 Global sizes of the whole distributed N-dimensional array.
 
NDArray localArray
 The N-dimensional array which stores the local elements and overlaps.
 
LocalBeginsType localBegins
 Global indices of the first local element in the whole N-dimensional array.
 
SizesHolderType localEnds
 Global indices of the end-of-range element, [localBegins, localEnds).
 

Detailed Description

template<typename NDArray>
class TNL::Containers::DistributedNDArray< NDArray >

Distributed N-dimensional array.

Template Parameters
NDArrayType of the N-dimensional array which is used to store the local elements. It can be NDArray or SlicedNDArray.
Example:
#include <iostream>
#include <TNL/Containers/BlockPartitioning.h>
#include <TNL/Containers/DistributedNDArray.h>
#include <TNL/Containers/DistributedNDArraySynchronizer.h>
#include <TNL/MPI/ScopedInitializer.h>
// The following works for any device (Host, Cuda, etc.)
template< typename Device >
void
distributedNDArrayExample()
{
using namespace TNL::Containers;
using LocalArrayType = NDArray< int, // Value
Device, // Device
int, // Index
>;
using LocalRangeType = typename ArrayType::LocalRangeType;
// set input parameters
const TNL::MPI::Comm communicator = MPI_COMM_WORLD;
const int num_rows = 10; // number of rows
const int num_cols = 4; // number of columns
constexpr int distributedAxis = 0; // 0: num_rows gets distributed, 1: num_cols gets distributed
// decompose the range of rows
const LocalRangeType localRange =
// create the distributed array
ArrayType a;
a.setSizes( num_rows, num_cols );
a.template setDistribution< distributedAxis >( localRange.getBegin(), localRange.getEnd(), communicator );
a.allocate();
// do some work with the array
auto a_view = a.getLocalView();
a.forAll(
[ = ] __cuda_callable__( int gi, int j ) mutable
{
// convert global row index to local
const int i = gi - localRange.getBegin();
// write the global row index to the array using local coordinates
a_view( i, j ) = gi;
} );
a.forGhosts(
[ = ] __cuda_callable__( int gi, int j ) mutable
{
// convert global row index to local
const int i = gi - localRange.getBegin();
// use the local indices to set ghost elements to -1
a_view( i, j ) = -1;
} );
// output the local elements as a flat array
ArrayView flat_view( a_view.getData(), a_view.getStorageSize() );
std::cout << "Rank " << communicator.rank() << " before synchronization: " << flat_view << std::endl;
// synchronize the ghost regions and output again
synchronizer.setSynchronizationPattern( NDArraySyncPatterns::D1Q3 );
synchronizer.synchronize( a );
std::cout << "Rank " << communicator.rank() << " after synchronization: " << flat_view << std::endl;
}
int
main( int argc, char* argv[] )
{
TNL::MPI::ScopedInitializer mpi( argc, argv );
if( TNL::MPI::GetRank() == 0 )
std::cout << "The first test runs on CPU ..." << std::endl;
distributedNDArrayExample< TNL::Devices::Host >();
#ifdef __CUDACC__
TNL::MPI::Barrier();
if( TNL::MPI::GetRank() == 0 )
std::cout << "The second test runs on GPU ..." << std::endl;
distributedNDArrayExample< TNL::Devices::Cuda >();
#endif
}
#define __cuda_callable__
Definition Macros.h:49
ArrayView is a simple data structure which provides a non-owning encapsulation of array data....
Definition ArrayView.h:55
Synchronizer for DistributedNDArray.
Definition DistributedNDArraySynchronizer.h:28
void setSynchronizationPattern(const std::array< SyncDirection, Q > &pattern)
Set the communication pattern between neighbors during data synchronization.
Definition DistributedNDArraySynchronizer.h:90
void synchronize(DistributedNDArray &array, SyncDirection mask=SyncDirection::All)
Synchronizes data in array distributed among MPI ranks.
Definition DistributedNDArraySynchronizer.h:199
Distributed N-dimensional array.
Definition DistributedNDArray.h:26
Dynamic N-dimensional array.
Definition NDArray.h:569
Holds static and dynamic sizes of an N-dimensional array.
Definition SizesHolder.h:30
Holds static sizes of an N-dimensional array.
Definition StaticSizesHolder.h:27
Definition Subrange.h:17
An RAII wrapper for custom MPI communicators.
Definition Comm.h:63
int rank() const
Determines the rank of the calling process in the communicator.
Definition Comm.h:216
T endl(T... args)
Namespace for TNL containers.
Definition Array.h:17
Subrange< Index > splitRange(Index rangeBegin, Index rangeEnd, int rank, int num_subintervals)
A helper function which splits a one-dimensional range.
Definition BlockPartitioning.h:27
Definition ScopedInitializer.h:63
Possible output:
Rank 0: rank on node is 0, using GPU id 0 of 2
Environment:
CUDA_VISIBLE_DEVICES=
Rank 1: rank on node is 1, using GPU id 1 of 2
Environment:
CUDA_VISIBLE_DEVICES=
Rank 2: rank on node is 2, using GPU id 0 of 2
Environment:
CUDA_VISIBLE_DEVICES=
Rank 3: rank on node is 3, using GPU id 1 of 2
Environment:
CUDA_VISIBLE_DEVICES=
Rank 3 before synchronization: [ -1, 8, 9, -1, -1, 8, 9, -1, -1, 8, 9, -1, -1, 8, 9, -1 ]
The first test runs on CPU ...
Rank 0 before synchronization: [ -1, 0, 1, 2, -1, -1, 0, 1, 2, -1, -1, 0, 1, 2, -1, -1, 0, 1, 2, -1 ]
Rank 2 before synchronization: [ -1, 6, 7, -1, -1, 6, 7, -1, -1, 6, 7, -1, -1, 6, 7, -1 ]
Rank 3 after synchronization: [ 7, 8, 9, 0, 7, 8, 9, 0, 7, 8, 9, 0, 7, 8, 9, 0 ]
Rank 1 before synchronization: [ -1, 3, 4, 5, -1, -1, 3, 4, 5, -1, -1, 3, 4, 5, -1, -1, 3, 4, 5, -1 ]
Rank 1 after synchronization: [ 2, 3, 4, 5, 6, 2, 3, 4, 5, 6, 2, 3, 4, 5, 6, 2, 3, 4, 5, 6 ]
Rank 0 after synchronization: [ 9, 0, 1, 2, 3, 9, 0, 1, 2, 3, 9, 0, 1, 2, 3, 9, 0, 1, 2, 3 ]
Rank 2 after synchronization: [ 5, 6, 7, 8, 5, 6, 7, 8, 5, 6, 7, 8, 5, 6, 7, 8 ]
The second test runs on GPU ...
Rank 3 before synchronization: [ -1, 8, 9, -1, -1, 8, 9, -1, -1, 8, 9, -1, -1, 8, 9, -1 ]
Rank 1 before synchronization: [ -1, 3, 4, 5, -1, -1, 3, 4, 5, -1, -1, 3, 4, 5, -1, -1, 3, 4, 5, -1 ]
Rank 0 before synchronization: [ -1, 0, 1, 2, -1, -1, 0, 1, 2, -1, -1, 0, 1, 2, -1, -1, 0, 1, 2, -1 ]
Rank 0 after synchronization: [ 9, 0, 1, 2, 3, 9, 0, 1, 2, 3, 9, 0, 1, 2, 3, 9, 0, 1, 2, 3 ]
Rank 2 before synchronization: [ -1, 6, 7, -1, -1, 6, 7, -1, -1, 6, 7, -1, -1, 6, 7, -1 ]
Rank 2 after synchronization: [ 5, 6, 7, 8, 5, 6, 7, 8, 5, 6, 7, 8, 5, 6, 7, 8 ]
Rank 3 after synchronization: [ 7, 8, 9, 0, 7, 8, 9, 0, 7, 8, 9, 0, 7, 8, 9, 0 ]
Rank 1 after synchronization: [ 2, 3, 4, 5, 6, 2, 3, 4, 5, 6, 2, 3, 4, 5, 6, 2, 3, 4, 5, 6 ]

Member Function Documentation

◆ forAll()

template<typename NDArray >
template<typename Device2 = DeviceType, typename Func >
void TNL::Containers::DistributedNDArray< NDArray >::forAll ( Func f,
const typename Device2::LaunchConfiguration & launch_configuration = typename Device2::LaunchConfiguration{} ) const
inline

Evaluates the function f in parallel for all elements of the array.

Each MPI rank iterates over all of its local elements.

See NDArrayView::forAll for the requirements on the function f.

◆ forBoundary() [1/2]

template<typename NDArray >
template<typename Device2 = DeviceType, typename SkipBegins , typename SkipEnds , typename Func >
void TNL::Containers::DistributedNDArray< NDArray >::forBoundary ( const SkipBegins & skipBegins,
const SkipEnds & skipEnds,
Func f,
const typename Device2::LaunchConfiguration & launch_configuration = typename Device2::LaunchConfiguration{} ) const
inline

Evaluates the function f in parallel for all elements outside the given [skipBegins, skipEnds) range specified by global indices.

Each MPI rank iterates over its local elements outside the range.

See NDArrayView::forAll for the requirements on the function f.

◆ forBoundary() [2/2]

template<typename NDArray >
template<typename Device2 = DeviceType, typename Func >
void TNL::Containers::DistributedNDArray< NDArray >::forBoundary ( Func f,
const typename Device2::LaunchConfiguration & launch_configuration = typename Device2::LaunchConfiguration{} ) const
inline

Evaluates the function f in parallel for all boundary elements of the array.

Each MPI rank iterates over its local elements which are neighbours of global boundaries.

See NDArrayView::forAll for the requirements on the function f.

◆ forGhosts()

template<typename NDArray >
template<typename Device2 = DeviceType, typename Func >
void TNL::Containers::DistributedNDArray< NDArray >::forGhosts ( Func f,
const typename Device2::LaunchConfiguration & launch_configuration = typename Device2::LaunchConfiguration{} ) const
inline

Evaluates the function f in parallel for all elements in the ghost region.

Each MPI rank iterates over elements which are in the overlapping region (i.e., owned by a different MPI rank). If all overlaps are 0, it has no effect.

See NDArrayView::forAll for the requirements on the function f.

◆ forInterior() [1/2]

template<typename NDArray >
template<typename Device2 = DeviceType, typename Begins , typename Ends , typename Func >
void TNL::Containers::DistributedNDArray< NDArray >::forInterior ( const Begins & begins,
const Ends & ends,
Func f,
const typename Device2::LaunchConfiguration & launch_configuration = typename Device2::LaunchConfiguration{} ) const
inline

Evaluates the function f in parallel for all elements inside the given [begins, ends) range specified by global indices.

Each MPI rank iterates over its local elements from the range.

See NDArrayView::forAll for the requirements on the function f.

◆ forInterior() [2/2]

template<typename NDArray >
template<typename Device2 = DeviceType, typename Func >
void TNL::Containers::DistributedNDArray< NDArray >::forInterior ( Func f,
const typename Device2::LaunchConfiguration & launch_configuration = typename Device2::LaunchConfiguration{} ) const
inline

Evaluates the function f in parallel for all internal elements of the array.

Each MPI rank iterates over its local elements which are not neighbours of global boundaries.

See NDArrayView::forAll for the requirements on the function f.

◆ forLocalBoundary()

template<typename NDArray >
template<typename Device2 = DeviceType, typename Func >
void TNL::Containers::DistributedNDArray< NDArray >::forLocalBoundary ( Func f,
const typename Device2::LaunchConfiguration & launch_configuration = typename Device2::LaunchConfiguration{} ) const
inline

Evaluates the function f in parallel for all local-boundary elements of the array.

Each MPI rank iterates over its local elements which are neighbours of overlaps. If all overlaps are 0, it has no effect.

See NDArrayView::forAll for the requirements on the function f.

◆ forLocalInterior()

template<typename NDArray >
template<typename Device2 = DeviceType, typename Func >
void TNL::Containers::DistributedNDArray< NDArray >::forLocalInterior ( Func f,
const typename Device2::LaunchConfiguration & launch_configuration = typename Device2::LaunchConfiguration{} ) const
inline

Evaluates the function f in parallel for all local-internal elements of the array.

Each MPI rank iterates over its local elements which are not neighbours of overlaps. If all overlaps are 0, it is equivalent to forAll.

See NDArrayView::forAll for the requirements on the function f.

◆ getElement()

template<typename NDArray >
template<typename... IndexTypes>
ValueType TNL::Containers::DistributedNDArray< NDArray >::getElement ( IndexTypes &&... indices) const
inlinenodiscard

A "safe" accessor for array elements.

It can be called only from the host code and it will do a slow copy from the device.

◆ getLocalRange()

template<typename NDArray >
template<std::size_t level>
LocalRangeType TNL::Containers::DistributedNDArray< NDArray >::getLocalRange ( ) const
inlinenodiscard

Returns a specific [begin, end) subrange of the local array in the global N-dimensional array.

Template Parameters
levelInteger specifying the component of the sizes to be returned.

◆ getOverlap()

template<typename NDArray >
template<std::size_t level>
__cuda_callable__ IndexType TNL::Containers::DistributedNDArray< NDArray >::getOverlap ( ) const
inlinenodiscard

Returns the overlap of a distributed N-dimensional array along the specified axis.

Template Parameters
levelInteger specifying the axis of the array.

◆ getSize()

template<typename NDArray >
template<std::size_t level>
IndexType TNL::Containers::DistributedNDArray< NDArray >::getSize ( ) const
inlinenodiscard

Returns a specific component of the N-dimensional global sizes.

Template Parameters
levelInteger specifying the component of the sizes to be returned.

◆ getStorageIndex()

template<typename NDArray >
template<typename... IndexTypes>
IndexType TNL::Containers::DistributedNDArray< NDArray >::getStorageIndex ( IndexTypes &&... indices) const
inlinenodiscard

Returns the local storage index for given global indices.

Parameters
indicesGlobal indices of the element in the N-dimensional array. The number of indices supplied must be equal to N, i.e. getDimension().
Returns
An index that can be used to address the element in a local one-dimensional array.

◆ operator()() [1/2]

template<typename NDArray >
template<typename... IndexTypes>
ValueType & TNL::Containers::DistributedNDArray< NDArray >::operator() ( IndexTypes &&... indices)
inlinenodiscard

Accesses an element of the array.

Only local elements or elements in the overlapping region can be accessed.

Parameters
indicesGlobal indices of the element in the N-dimensional array. The number of indices supplied must be equal to N, i.e. getDimension().
Returns
Reference to the array element.

◆ operator()() [2/2]

template<typename NDArray >
template<typename... IndexTypes>
const ValueType & TNL::Containers::DistributedNDArray< NDArray >::operator() ( IndexTypes &&... indices) const
inlinenodiscard

Accesses an element of the array.

Only local elements or elements in the overlapping region can be accessed.

Parameters
indicesGlobal indices of the element in the N-dimensional array. The number of indices supplied must be equal to N, i.e. getDimension().
Returns
Constant reference to the array element.

◆ operator[]() [1/2]

template<typename NDArray >
ValueType & TNL::Containers::DistributedNDArray< NDArray >::operator[] ( IndexType index)
inlinenodiscard

Accesses an element in a one-dimensional array.

Only local elements or elements in the overlapping region can be accessed.

Warning
This function can be used only when the dimension of the array is equal to 1.
Parameters
indexGlobal index of the element in the one-dimensional array.
Returns
Reference to the array element.

◆ operator[]() [2/2]

template<typename NDArray >
const ValueType & TNL::Containers::DistributedNDArray< NDArray >::operator[] ( IndexType index) const
inlinenodiscard

Accesses an element in a one-dimensional array.

Only local elements or elements in the overlapping region can be accessed.

Warning
This function can be used only when the dimension of the array is equal to 1.
Parameters
indexGlobal index of the element in the one-dimensional array.
Returns
Reference to the array element.

◆ reset()

template<typename NDArray >
void TNL::Containers::DistributedNDArray< NDArray >::reset ( )
inline

Resets the array to the empty state.

The current data will be deallocated, thus all pointers and views to the array elements will become invalid.

◆ setDistribution()

template<typename NDArray >
template<std::size_t level>
void TNL::Containers::DistributedNDArray< NDArray >::setDistribution ( IndexType begin,
IndexType end,
const MPI::Comm & communicator = MPI_COMM_WORLD )
inline

Sets the distribution of the arrray among MPI ranks along the given axis.

Template Parameters
levelInteger specifying the axis of the array to be distributed.
Parameters
beginIndex of the first element owned by the calling MPI rank.
endIndex of the first element not owned by the calling MPI rank. I.e., the calling rank owns the range [begin, end).
communicatorMPI communicator to associate with the array.

◆ setLike()

template<typename NDArray >
void TNL::Containers::DistributedNDArray< NDArray >::setLike ( const DistributedNDArray< NDArray > & other)
inline

Sets sizes of the array to the sizes of an existing array.

If the array size changes, the current data will be deallocated, thus all pointers and views to the array alements will become invalid.

Member Data Documentation

◆ localBegins

template<typename NDArray >
LocalBeginsType TNL::Containers::DistributedNDArray< NDArray >::localBegins
protected

Global indices of the first local element in the whole N-dimensional array.

Note that localBegins and localEnds have different static sizes (and hence different C++ type): localBegins is always 0, localEnds has always the full static size.


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