Distributed operator

Builder

template<typename CoefficientPrecision, typename CoordinatePrecision = underlying_type<CoefficientPrecision>>
class DefaultApproximationBuilder

Public Functions

inline DefaultApproximationBuilder(const VirtualInternalGenerator<CoefficientPrecision> &generator, const Cluster<CoordinatePrecision> &target_cluster, const Cluster<CoordinatePrecision> &source_cluster, const HMatrixTreeBuilder<CoefficientPrecision, CoordinatePrecision> &hmatrix_tree_builder, MPI_Comm communicator)

inline DefaultApproximationBuilder(const VirtualGenerator<CoefficientPrecision> &generator, const Cluster<CoordinatePrecision> &target_cluster, const Cluster<CoordinatePrecision> &source_cluster, const HMatrixTreeBuilder<CoefficientPrecision, CoordinatePrecision> &hmatrix_tree_builder, MPI_Comm communicator)

Public Members

HMatrix<CoefficientPrecision, CoordinatePrecision> hmatrix

DistributedOperator<CoefficientPrecision> &distributed_operator

const HMatrix<CoefficientPrecision, CoordinatePrecision> *block_diagonal_hmatrix = {nullptr}

template<typename CoefficientPrecision, typename CoordinatePrecision = underlying_type<CoefficientPrecision>>
class DefaultLocalApproximationBuilder

Public Functions

inline DefaultLocalApproximationBuilder(const VirtualInternalGenerator<CoefficientPrecision> &generator, const Cluster<CoordinatePrecision> &target_cluster, const Cluster<CoordinatePrecision> &source_cluster, const HMatrixTreeBuilder<CoefficientPrecision, CoordinatePrecision> &hmatrix_tree_builder, MPI_Comm communicator)

inline DefaultLocalApproximationBuilder(const VirtualGenerator<CoefficientPrecision> &generator, const Cluster<CoordinatePrecision> &target_cluster, const Cluster<CoordinatePrecision> &source_cluster, const HMatrixTreeBuilder<CoefficientPrecision, CoordinatePrecision> &hmatrix_tree_builder, MPI_Comm communicator)

Public Members

HMatrix<CoefficientPrecision, CoordinatePrecision> hmatrix

DistributedOperator<CoefficientPrecision> &distributed_operator

const HMatrix<CoefficientPrecision, CoordinatePrecision> *block_diagonal_hmatrix = {nullptr}

template<typename CoefficientPrecision, typename CoordinatePrecision = underlying_type<CoefficientPrecision>>
class CustomApproximationBuilder

Public Functions

inline explicit CustomApproximationBuilder(const Cluster<CoordinatePrecision> &target_cluster, const Cluster<CoordinatePrecision> &source_cluster, char symmetry, char UPLO, MPI_Comm communicator, const VirtualLocalOperator<CoefficientPrecision> &local_operator)

Public Members

DistributedOperator<CoefficientPrecision> distributed_operator

DistributedOperator

template<typename CoefficientPrecision>
class DistributedOperator

Public Functions

DistributedOperator(const DistributedOperator&) = delete

DistributedOperator &operator=(const DistributedOperator&) = delete

DistributedOperator(DistributedOperator &&cluster) noexcept = default

DistributedOperator &operator=(DistributedOperator &&cluster) noexcept = default

virtual ~DistributedOperator() = default

inline explicit DistributedOperator(const VirtualPartition<CoefficientPrecision> &target_partition, const VirtualPartition<CoefficientPrecision> &source_partition, char symmetry, char UPLO, MPI_Comm comm)

inline void add_local_operator(const VirtualLocalOperator<CoefficientPrecision> *local_operator)

void vector_product_global_to_global(const CoefficientPrecision *const in, CoefficientPrecision *const out) const

void matrix_product_global_to_global(const CoefficientPrecision *const in, CoefficientPrecision *const out, int mu) const

void vector_product_transp_global_to_global(const CoefficientPrecision *const in, CoefficientPrecision *const out) const

void matrix_product_transp_global_to_global(const CoefficientPrecision *const in, CoefficientPrecision *const out, int mu) const

void vector_product_local_to_local(const CoefficientPrecision *const in, CoefficientPrecision *const out, CoefficientPrecision *work = nullptr) const

void matrix_product_local_to_local(const CoefficientPrecision *const in, CoefficientPrecision *const out, int mu, CoefficientPrecision *work = nullptr) const

void vector_product_transp_local_to_local(const CoefficientPrecision *const in, CoefficientPrecision *const out, CoefficientPrecision *work = nullptr) const

void matrix_product_transp_local_to_local(const CoefficientPrecision *const in, CoefficientPrecision *const out, int mu, CoefficientPrecision *work = nullptr) const

void internal_vector_product_global_to_local(const CoefficientPrecision *const in, CoefficientPrecision *const out) const

void internal_matrix_product_global_to_local(const CoefficientPrecision *const in, CoefficientPrecision *const out, int mu) const

void internal_vector_product_transp_local_to_global(const CoefficientPrecision *const in, CoefficientPrecision *const out) const

void internal_matrix_product_transp_local_to_global(const CoefficientPrecision *const in, CoefficientPrecision *const out, int mu) const

void internal_vector_product_local_to_local(const CoefficientPrecision *const in, CoefficientPrecision *const out, CoefficientPrecision *work = nullptr) const

void internal_matrix_product_local_to_local(const CoefficientPrecision *const in, CoefficientPrecision *const out, int mu, CoefficientPrecision *work = nullptr) const

void internal_vector_product_transp_local_to_local(const CoefficientPrecision *const in, CoefficientPrecision *const out, CoefficientPrecision *work = nullptr) const

void internal_matrix_product_transp_local_to_local(const CoefficientPrecision *const in, CoefficientPrecision *const out, int mu, CoefficientPrecision *work = nullptr) const

void internal_sub_matrix_product_to_local(const CoefficientPrecision *const in, CoefficientPrecision *const out, int mu, int offset, int size) const

void local_to_global_source(const CoefficientPrecision *const in, CoefficientPrecision *const out, const int &mu) const

void local_to_global_target(const CoefficientPrecision *const in, CoefficientPrecision *const out, const int &mu) const

inline bool &use_permutation()

inline const bool &use_permutation() const

inline char get_symmetry_type() const

inline char get_storage_type() const

inline MPI_Comm get_comm() const

inline const VirtualPartition<CoefficientPrecision> &get_target_partition() const

inline const VirtualPartition<CoefficientPrecision> &get_source_partition() const