|
using | ByteArrayView = typename Base::ByteArrayView |
|
using | DeviceType = typename DistributedMesh::DeviceType |
|
using | GlobalIndexType = typename DistributedMesh::GlobalIndexType |
|
using | RequestsVector = typename Base::RequestsVector |
|
enum | AsyncPolicy |
|
using | ByteArrayView |
|
using | RequestsVector |
|
|
int | getEntityOwner (GlobalIndexType global_idx) const |
|
const auto & | getGhostEntitiesCounts () const |
|
const auto & | getGhostNeighborOffsets () const |
|
const auto & | getGhostNeighbors () const |
|
const auto & | getGhostOffsets () const |
|
const auto & | getGlobalOffsets () const |
|
void | initialize (const DistributedMesh &mesh) |
|
template<typename Array, std::enable_if_t< ! HasMeshType< Array >::value, bool > = true> |
void | synchronize (Array &array) |
|
template<typename MeshFunction, std::enable_if_t< HasMeshType< MeshFunction >::value, bool > = true> |
void | synchronize (MeshFunction &function) |
|
template<typename Array> |
void | synchronizeArray (Array &array, int valuesPerElement=1) |
|
void | synchronizeByteArray (ByteArrayView array, int bytesPerValue) override |
| Main synchronization function.
|
|
RequestsVector | synchronizeByteArrayAsyncWorker (ByteArrayView array, int bytesPerValue) override |
|
template<typename SparsePattern> |
auto | synchronizeSparse (const SparsePattern &pattern, bool assumeConsistentRowCapacities=false) |
|
void | synchronizeByteArrayAsync (ByteArrayView array, int bytesPerValue, AsyncPolicy policy=AsyncPolicy::synchronous) |
| An asynchronous version of synchronizeByteArray.
|
|
|
MPI::Comm | communicator |
|
Matrices::DenseMatrix< GlobalIndexType, Devices::Host, int > | ghostEntitiesCounts |
|
Containers::Array< GlobalIndexType, Devices::Host, int > | ghostNeighborOffsets |
|
Containers::Vector< GlobalIndexType, DeviceType, GlobalIndexType > | ghostNeighbors |
|
Containers::Array< GlobalIndexType, Devices::Host, int > | ghostOffsets |
|
Containers::Array< GlobalIndexType, Devices::Host, int > | globalOffsets |
|
Containers::Array< std::uint8_t, DeviceType, GlobalIndexType > | sendBuffers |
|
|
std::future< void > | async_op |
| Can be used for checking if a synchronization started asynchronously has been finished.
|
|
std::size_t | async_ops_count |
|
Timer | async_start_timer |
|
Timer | async_wait_before_start_timer |
|
Timer | async_wait_timer |
|
◆ synchronizeByteArray()
template<typename
DistributedMesh, int EntityDimension = DistributedMesh::getMeshDimension()>
◆ synchronizeByteArrayAsyncWorker()
template<typename
DistributedMesh, int EntityDimension = DistributedMesh::getMeshDimension()>
|
inlinenodiscardoverridevirtual |
◆ ghostEntitiesCounts
template<typename
DistributedMesh, int EntityDimension = DistributedMesh::getMeshDimension()>
Communication pattern:
- an unsymmetric nproc x nproc matrix G such that G_ij represents the number of ghost entities on rank i that are owned by rank j
- assembly of the i-th row involves traversal of the ghost entities on the local mesh and determining its owner based on the global index
- assembly of the full matrix needs all-to-all communication
- for the i-th rank, the i-th row determines the receive buffer sizes and the i-th column determines the send buffer sizes
◆ ghostNeighborOffsets
template<typename
DistributedMesh, int EntityDimension = DistributedMesh::getMeshDimension()>
Ghost neighbor offsets: array of size nproc + 1 where the i-th value is the offset of ghost neighbor indices requested by the i-th rank. The last value is the size of the ghostNeighbors and sendBuffers arrays (see below).
◆ ghostNeighbors
template<typename
DistributedMesh, int EntityDimension = DistributedMesh::getMeshDimension()>
Ghost neighbor indices: array containing local indices of the entities which are ghosts on other ranks. The indices requested by the i-th rank are in the range starting at ghostNeighborOffsets[i] (inclusive) and ending at ghostNeighborOffsets[i+1] (exclusive). These indices are used for copying the mesh function values into the sendBuffers array. Note that ghost neighbor indices cannot be made contiguous in general so we need the send buffers.
◆ ghostOffsets
template<typename
DistributedMesh, int EntityDimension = DistributedMesh::getMeshDimension()>
Ghost offsets: array of size nproc + 1 where the i-th value is the local index of the first ghost entity owned by the i-th rank. The last value is equal to the entities count on the local mesh. All ghost entities owned by the i-th rank are assumed to be indexed contiguously in the local mesh.
◆ globalOffsets
template<typename
DistributedMesh, int EntityDimension = DistributedMesh::getMeshDimension()>
Global offsets: array of size nproc where the i-th value is the lowest global index of the entities owned by the i-th rank. This can be used to determine the owner of every entity based on its global index.
◆ sendBuffers
template<typename
DistributedMesh, int EntityDimension = DistributedMesh::getMeshDimension()>
Send buffers: array for buffering the mesh function values which will be sent to other ranks. We use std::uint8_t as the value type to make this class independent of the mesh function's real type. When cast to the real type values, the send buffer for the i-th rank is the part of the array starting at index ghostNeighborOffsets[i] (inclusive) and ending at index ghostNeighborOffsets[i+1] (exclusive).
The documentation for this class was generated from the following file:
- src/TNL/Meshes/DistributedMeshes/DistributedMeshSynchronizer.h