The space adapted for Trilinos vectors and matrices. More...

#include <trilinos_space.h>

Collaboration diagram for Kratos::TrilinosSpace< TMatrixType, TVectorType >:

Public Member Functions
Life Cycle
	TrilinosSpace ()
	Default constructor. More...

virtual	~TrilinosSpace ()
	Destructor. More...

Type Definitions
using	DataType = double
	Definition of the data type. More...

using	MatrixType = TMatrixType
	Definition of the matrix type. More...

using	VectorType = TVectorType
	Definition of the vector type. More...

using	IndexType = std::size_t
	Definition of the index type. More...

using	SizeType = std::size_t
	Definition of the size type. More...

using	MatrixPointerType = typename Kratos::shared_ptr< TMatrixType >
	Definition of the pointer types. More...

using	VectorPointerType = typename Kratos::shared_ptr< TVectorType >

using	DofUpdaterType = TrilinosDofUpdater< TrilinosSpace< TMatrixType, TVectorType > >
	Some other definitions. More...

using	DofUpdaterPointerType = typename DofUpdater< TrilinosSpace< TMatrixType, TVectorType > >::UniquePointer

	KRATOS_CLASS_POINTER_DEFINITION (TrilinosSpace)
	Pointer definition of TrilinosSpace. More...

Operations
MatrixPointerType	ReadMatrixMarket (const std::string FileName, Epetra_MpiComm &rComm)
	Read a matrix from a MatrixMarket file. More...

VectorPointerType	ReadMatrixMarketVector (const std::string &rFileName, Epetra_MpiComm &rComm, const int N)
	Read a vector from a MatrixMarket file. More...

static MatrixPointerType	CreateEmptyMatrixPointer ()
	This method creates an empty pointer to a matrix. More...

static VectorPointerType	CreateEmptyVectorPointer ()
	This method creates an empty pointer to a vector. More...

static MatrixPointerType	CreateEmptyMatrixPointer (Epetra_MpiComm &rComm)
	This method creates an empty pointer to a matrix using epetra communicator. More...

static VectorPointerType	CreateEmptyVectorPointer (Epetra_MpiComm &rComm)
	This method creates an empty pointer to a vector using epetra communicator. More...

static IndexType	Size (const VectorType &rV)
	Returns size of vector rV. More...

static IndexType	Size1 (MatrixType const &rM)
	Returns number of rows of rM. More...

static IndexType	Size2 (MatrixType const &rM)
	Returns number of columns of rM. More...

static void	GetColumn (const unsigned int j, const MatrixType &rM, VectorType &rX)
	Returns the column of the matrix in the given position. More...

static void	Copy (const MatrixType &rX, MatrixType &rY)
	Returns a copy of the matrix rX. More...

static void	Copy (const VectorType &rX, VectorType &rY)
	Returns a copy of the vector rX. More...

static double	Dot (const VectorType &rX, const VectorType &rY)
	Returns the product of two vectors. More...

static double	Max (const VectorType &rX)
	Returns the maximum value of the vector rX. More...

static double	Min (const VectorType &rX)
	Returns the minimum value of the vector rX. More...

static double	TwoNorm (const VectorType &rX)
	Returns the norm of the vector rX. More...

static double	TwoNorm (const MatrixType &rA)
	Returns the Frobenius norm of the matrix rX. More...

static void	Mult (const MatrixType &rA, const VectorType &rX, VectorType &rY)
	Returns the multiplication of a matrix by a vector. More...

static void	Mult (const MatrixType &rA, const MatrixType &rB, MatrixType &rC, const bool CallFillCompleteOnResult=true, const bool KeepAllHardZeros=false)
	Returns the multiplication matrix-matrix. More...

static void	TransposeMult (const MatrixType &rA, const VectorType &rX, VectorType &rY)
	Returns the transpose multiplication of a matrix by a vector. More...

static void	TransposeMult (const MatrixType &rA, const MatrixType &rB, MatrixType &rC, const std::pair< bool, bool > TransposeFlag={false, false}, const bool CallFillCompleteOnResult=true, const bool KeepAllHardZeros=false)
	Returns the transpose multiplication matrix-matrix. More...

static void	BtDBProductOperation (MatrixType &rA, const MatrixType &rD, const MatrixType &rB, const bool CallFillCompleteOnResult=true, const bool KeepAllHardZeros=false, const bool EnforceInitialGraph=false)
	Calculates the product operation B'DB. More...

static void	BDBtProductOperation (MatrixType &rA, const MatrixType &rD, const MatrixType &rB, const bool CallFillCompleteOnResult=true, const bool KeepAllHardZeros=false, const bool EnforceInitialGraph=false)
	Calculates the product operation BDB'. More...

static void	InplaceMult (VectorType &rX, const double A)
	Returns the multiplication of a vector by a scalar. More...

static void	Assign (VectorType &rX, const double A, const VectorType &rY)
	Returns the multiplication of a vector by a scalar. More...

static void	UnaliasedAdd (VectorType &rX, const double A, const VectorType &rY)
	Returns the unaliased addition of a vector by a scalar times a vector. More...

static void	ScaleAndAdd (const double A, const VectorType &rX, const double B, const VectorType &rY, VectorType &rZ)
	Returns the unaliased addition of two vectors by a scalar. More...

static void	ScaleAndAdd (const double A, const VectorType &rX, const double B, VectorType &rY)
	Returns the unaliased addition of two vectors by a scalar. More...

static void	SetValue (VectorType &rX, IndexType i, const double value)
	Sets a value in a vector. More...

static void	Set (VectorType &rX, const DataType A)
	assigns a scalar to a vector More...

static void	Resize (MatrixType &rA, const SizeType m, const SizeType n)
	Resizes a matrix. More...

static void	Resize (VectorType &rX, const SizeType n)
	Resizes a vector. More...

static void	Resize (VectorPointerType &pX, const SizeType n)
	Resizes a vector. More...

static void	Clear (MatrixPointerType &pA)
	Clears a matrix. More...

static void	Clear (VectorPointerType &pX)
	Clears a vector. More...

static void	SetToZero (MatrixType &rA)
	Sets a matrix to zero. More...

static void	SetToZero (VectorType &rX)
	Sets a vector to zero. More...

static void	AssembleLHS (MatrixType &rA, const Matrix &rLHSContribution, const std::vector< std::size_t > &rEquationId)
	TODO: creating the the calculating reaction version. More...

static void	AssembleRHS (VectorType &rb, const Vector &rRHSContribution, const std::vector< std::size_t > &rEquationId)
	TODO: creating the the calculating reaction version. More...

static constexpr bool	IsDistributed ()
	This function returns if we are in a distributed system. More...

static double	GetValue (const VectorType &rX, const std::size_t I)
	This function returns a value from a given vector according to a given index. More...

static void	GatherValues (const VectorType &rX, const std::vector< int > &IndexArray, double *pValues)
	This function gathers the values of a given vector according to a given index array. More...

static Epetra_CrsGraph	CombineMatricesGraphs (const MatrixType &rA, const MatrixType &rB)
	Generates a graph combining the graphs of two matrices. More...

static void	CopyMatrixValues (MatrixType &rA, const MatrixType &rB)
	Copy values from one matrix to another. More...

static double	CheckAndCorrectZeroDiagonalValues (const ProcessInfo &rProcessInfo, MatrixType &rA, VectorType &rb, const SCALING_DIAGONAL ScalingDiagonal=SCALING_DIAGONAL::NO_SCALING)
	This method checks and corrects the zero diagonal values. More...

static double	GetScaleNorm (const ProcessInfo &rProcessInfo, const MatrixType &rA, const SCALING_DIAGONAL ScalingDiagonal=SCALING_DIAGONAL::NO_SCALING)
	This method returns the scale norm considering for scaling the diagonal. More...

static double	GetDiagonalNorm (const MatrixType &rA)
	This method returns the diagonal norm considering for scaling the diagonal. More...

static double	GetAveragevalueDiagonal (const MatrixType &rA)
	This method returns the diagonal max value. More...

static double	GetMaxDiagonal (const MatrixType &rA)
	This method returns the diagonal max value. More...

static double	GetMinDiagonal (const MatrixType &rA)
	This method returns the diagonal min value. More...

static constexpr bool	IsDistributedSpace ()
	Check if the TrilinosSpace is distributed. More...

Input and output
virtual std::string	Info () const
	Turn back information as a string. More...

virtual void	PrintInfo (std::ostream &rOStream) const
	Print information about this object. More...

virtual void	PrintData (std::ostream &rOStream) const
	Print object's data. More...

template<class TOtherMatrixType >
static bool	WriteMatrixMarketMatrix (const char *pFileName, const TOtherMatrixType &rM, const bool Symmetric)
	Writes a matrix to a file in MatrixMarket format. More...

template<class VectorType >
static bool	WriteMatrixMarketVector (const char *pFileName, const VectorType &rV)
	Writes a vector to a file in MatrixMarket format. More...

static DofUpdaterPointerType	CreateDofUpdater ()
	Creates a new dof updater. More...

Detailed Description

template<class TMatrixType, class TVectorType>
class Kratos::TrilinosSpace< TMatrixType, TVectorType >

The space adapted for Trilinos vectors and matrices.

Author: Riccardo Rossi

Template Parameters

TMatrixType	The matrix type considered
TVectorType	the vector type considered

Member Typedef Documentation

◆ DataType

template<class TMatrixType , class TVectorType >

using Kratos::TrilinosSpace< TMatrixType, TVectorType >::DataType = double

Definition of the data type.

◆ DofUpdaterPointerType

template<class TMatrixType , class TVectorType >

using Kratos::TrilinosSpace< TMatrixType, TVectorType >::DofUpdaterPointerType = typename DofUpdater<TrilinosSpace<TMatrixType,TVectorType> >::UniquePointer

◆ DofUpdaterType

template<class TMatrixType , class TVectorType >

using Kratos::TrilinosSpace< TMatrixType, TVectorType >::DofUpdaterType = TrilinosDofUpdater< TrilinosSpace<TMatrixType,TVectorType> >

Some other definitions.

◆ IndexType

template<class TMatrixType , class TVectorType >

using Kratos::TrilinosSpace< TMatrixType, TVectorType >::IndexType = std::size_t

Definition of the index type.

◆ MatrixPointerType

template<class TMatrixType , class TVectorType >

using Kratos::TrilinosSpace< TMatrixType, TVectorType >::MatrixPointerType = typename Kratos::shared_ptr<TMatrixType>

Definition of the pointer types.

◆ MatrixType

template<class TMatrixType , class TVectorType >

using Kratos::TrilinosSpace< TMatrixType, TVectorType >::MatrixType = TMatrixType

Definition of the matrix type.

◆ SizeType

template<class TMatrixType , class TVectorType >

using Kratos::TrilinosSpace< TMatrixType, TVectorType >::SizeType = std::size_t

Definition of the size type.

◆ VectorPointerType

template<class TMatrixType , class TVectorType >

using Kratos::TrilinosSpace< TMatrixType, TVectorType >::VectorPointerType = typename Kratos::shared_ptr<TVectorType>

◆ VectorType

template<class TMatrixType , class TVectorType >

using Kratos::TrilinosSpace< TMatrixType, TVectorType >::VectorType = TVectorType

Definition of the vector type.

Constructor & Destructor Documentation

◆ TrilinosSpace()

template<class TMatrixType , class TVectorType >

Kratos::TrilinosSpace< TMatrixType, TVectorType >::TrilinosSpace ( )

inline

Default constructor.

◆ ~TrilinosSpace()

template<class TMatrixType , class TVectorType >

virtual Kratos::TrilinosSpace< TMatrixType, TVectorType >::~TrilinosSpace ( )

inlinevirtual

Destructor.

Member Function Documentation

◆ AssembleLHS()

template<class TMatrixType , class TVectorType >

static void Kratos::TrilinosSpace< TMatrixType, TVectorType >::AssembleLHS	(	MatrixType &	rA,
		const Matrix &	rLHSContribution,
		const std::vector< std::size_t > &	rEquationId
	)

inlinestatic

TODO: creating the the calculating reaction version.

Assembles the LHS of the system

Parameters

rA	The LHS matrix
rLHSContribution	The contribution to the LHS
rEquationId	The equation ids

◆ AssembleRHS()

template<class TMatrixType , class TVectorType >

static void Kratos::TrilinosSpace< TMatrixType, TVectorType >::AssembleRHS	(	VectorType &	rb,
		const Vector &	rRHSContribution,
		const std::vector< std::size_t > &	rEquationId
	)

inlinestatic

TODO: creating the the calculating reaction version.

Assembles the RHS of the system

Parameters

rb	The RHS vector
rRHSContribution	The RHS contribution
rEquationId	The equation ids

◆ Assign()

template<class TMatrixType , class TVectorType >

static void Kratos::TrilinosSpace< TMatrixType, TVectorType >::Assign	(	VectorType &	rX,
		const double	A,
		const VectorType &	rY
	)

inlinestatic

Returns the multiplication of a vector by a scalar.

x = A*y Checks if a multiplication is needed and tries to do otherwise

Note: ATTENTION it is assumed no aliasing between rX and rY

Parameters

rX	The resulting vector considered
A	The scalar considered
rY	The multiplied vector considered

◆ BDBtProductOperation()

template<class TMatrixType , class TVectorType >

static void Kratos::TrilinosSpace< TMatrixType, TVectorType >::BDBtProductOperation	(	MatrixType &	rA,
		const MatrixType &	rD,
		const MatrixType &	rB,
		const bool	CallFillCompleteOnResult = `true`,
		const bool	KeepAllHardZeros = `false`,
		const bool	EnforceInitialGraph = `false`
	)

inlinestatic

Calculates the product operation BDB'.

Parameters

rA	The resulting matrix
rD	The "center" matrix
rB	The matrices to be transposed
CallFillCompleteOnResult	Optional argument, defaults to true. Power users may specify this argument to be false if they DON'T want this function to call C.FillComplete. (It is often useful to allow this function to call C.FillComplete, in cases where one or both of the input matrices are rectangular and it is not trivial to know which maps to use for the domain- and range-maps.)
KeepAllHardZeros	Optional argument, defaults to false. If true, Multiply, keeps all entries in C corresponding to hard zeros. If false, the following happens by case: AB^T, A^TB^T - Does not store entries caused by hard zeros in C. A^TB (unoptimized) - Hard zeros are always stored (this option has no effect) AB, A^T*B (optimized) - Hard zeros in corresponding to hard zeros in A are not stored, There are certain cases involving reuse of C, where this can be useful.

Template Parameters

TEnforceInitialGraph If the initial graph is enforced, or a new one is generated

◆ BtDBProductOperation()

template<class TMatrixType , class TVectorType >

static void Kratos::TrilinosSpace< TMatrixType, TVectorType >::BtDBProductOperation	(	MatrixType &	rA,
		const MatrixType &	rD,
		const MatrixType &	rB,
		const bool	CallFillCompleteOnResult = `true`,
		const bool	KeepAllHardZeros = `false`,
		const bool	EnforceInitialGraph = `false`
	)

inlinestatic

Calculates the product operation B'DB.

Parameters

rA	The resulting matrix
rD	The "center" matrix
rB	The matrices to be transposed
CallFillCompleteOnResult	Optional argument, defaults to true. Power users may specify this argument to be false if they DON'T want this function to call C.FillComplete. (It is often useful to allow this function to call C.FillComplete, in cases where one or both of the input matrices are rectangular and it is not trivial to know which maps to use for the domain- and range-maps.)
KeepAllHardZeros	Optional argument, defaults to false. If true, Multiply, keeps all entries in C corresponding to hard zeros. If false, the following happens by case: AB^T, A^TB^T - Does not store entries caused by hard zeros in C. A^TB (unoptimized) - Hard zeros are always stored (this option has no effect) AB, A^T*B (optimized) - Hard zeros in corresponding to hard zeros in A are not stored, There are certain cases involving reuse of C, where this can be useful.

Template Parameters

TEnforceInitialGraph If the initial graph is enforced, or a new one is generated

◆ CheckAndCorrectZeroDiagonalValues()

template<class TMatrixType , class TVectorType >

static double Kratos::TrilinosSpace< TMatrixType, TVectorType >::CheckAndCorrectZeroDiagonalValues	(	const ProcessInfo &	rProcessInfo,
		MatrixType &	rA,
		VectorType &	rb,
		const SCALING_DIAGONAL	ScalingDiagonal = `SCALING_DIAGONAL::NO_SCALING`
	)

inlinestatic

This method checks and corrects the zero diagonal values.

This method returns the scale norm considering for scaling the diagonal

Parameters

rProcessInfo	The problem process info
rA	The LHS matrix
rb	The RHS vector
ScalingDiagonal	The type of caling diagonal considered

Returns: The scale norm

◆ Clear() [1/2]

template<class TMatrixType , class TVectorType >

static void Kratos::TrilinosSpace< TMatrixType, TVectorType >::Clear ( MatrixPointerType & pA )

inlinestatic

Clears a matrix.

Parameters

pA	The pointer to the matrix to be cleared

◆ Clear() [2/2]

template<class TMatrixType , class TVectorType >

static void Kratos::TrilinosSpace< TMatrixType, TVectorType >::Clear ( VectorPointerType & pX )

inlinestatic

Clears a vector.

Parameters

pX	The pointer to the vector to be cleared

◆ CombineMatricesGraphs()

template<class TMatrixType , class TVectorType >

static Epetra_CrsGraph Kratos::TrilinosSpace< TMatrixType, TVectorType >::CombineMatricesGraphs	(	const MatrixType &	rA,
		const MatrixType &	rB
	)

inlinestatic

Generates a graph combining the graphs of two matrices.

Parameters

rA	The first matrix
rB	The second matrix

◆ Copy() [1/2]

template<class TMatrixType , class TVectorType >

static void Kratos::TrilinosSpace< TMatrixType, TVectorType >::Copy	(	const MatrixType &	rX,
		MatrixType &	rY
	)

inlinestatic

Returns a copy of the matrix rX.

rY = rX

Parameters

rX	The matrix considered
rY	The copy of the matrix rX

◆ Copy() [2/2]

template<class TMatrixType , class TVectorType >

static void Kratos::TrilinosSpace< TMatrixType, TVectorType >::Copy	(	const VectorType &	rX,
		VectorType &	rY
	)

inlinestatic

Returns a copy of the vector rX.

rY = rX

Parameters

rX	The vector considered
rY	The copy of the vector rX

◆ CopyMatrixValues()

template<class TMatrixType , class TVectorType >

static void Kratos::TrilinosSpace< TMatrixType, TVectorType >::CopyMatrixValues	(	MatrixType &	rA,
		const MatrixType &	rB
	)

inlinestatic

Copy values from one matrix to another.

It is assumed that the sparcity of both matrices is compatible

Parameters

rA	The matrix where assigning values
rB	The matrix to be copied

◆ CreateDofUpdater()

template<class TMatrixType , class TVectorType >

static DofUpdaterPointerType Kratos::TrilinosSpace< TMatrixType, TVectorType >::CreateDofUpdater ( )

inlinestatic

Creates a new dof updater.

Returns: The new dof updater

◆ CreateEmptyMatrixPointer() [1/2]

template<class TMatrixType , class TVectorType >

static MatrixPointerType Kratos::TrilinosSpace< TMatrixType, TVectorType >::CreateEmptyMatrixPointer ( )

inlinestatic

This method creates an empty pointer to a matrix.

Returns: The pointer to the matrix

◆ CreateEmptyMatrixPointer() [2/2]

template<class TMatrixType , class TVectorType >

static MatrixPointerType Kratos::TrilinosSpace< TMatrixType, TVectorType >::CreateEmptyMatrixPointer ( Epetra_MpiComm & rComm )

inlinestatic

This method creates an empty pointer to a matrix using epetra communicator.

Parameters

rComm The epetra communicator

Returns: The pointer to the matrix

◆ CreateEmptyVectorPointer() [1/2]

template<class TMatrixType , class TVectorType >

static VectorPointerType Kratos::TrilinosSpace< TMatrixType, TVectorType >::CreateEmptyVectorPointer ( )

inlinestatic

This method creates an empty pointer to a vector.

Returns: The pointer to the vector

◆ CreateEmptyVectorPointer() [2/2]

template<class TMatrixType , class TVectorType >

static VectorPointerType Kratos::TrilinosSpace< TMatrixType, TVectorType >::CreateEmptyVectorPointer ( Epetra_MpiComm & rComm )

inlinestatic

This method creates an empty pointer to a vector using epetra communicator.

Parameters

rComm The epetra communicator

Returns: The pointer to the vector

◆ Dot()

template<class TMatrixType , class TVectorType >

static double Kratos::TrilinosSpace< TMatrixType, TVectorType >::Dot	(	const VectorType &	rX,
		const VectorType &	rY
	)

inlinestatic

Returns the product of two vectors.

rX * rY

Parameters

rX	The first vector considered
rY	The second vector considered

◆ GatherValues()

template<class TMatrixType , class TVectorType >

static void Kratos::TrilinosSpace< TMatrixType, TVectorType >::GatherValues	(	const VectorType &	rX,
		const std::vector< int > &	IndexArray,
		double *	pValues
	)

inlinestatic

This function gathers the values of a given vector according to a given index array.

Parameters

rX	The vector from which values are to be gathered
IndexArray	The array containing the indices of the values to be gathered
pValues	The array containing the gathered values

◆ GetAveragevalueDiagonal()

template<class TMatrixType , class TVectorType >

static double Kratos::TrilinosSpace< TMatrixType, TVectorType >::GetAveragevalueDiagonal ( const MatrixType & rA )

inlinestatic

This method returns the diagonal max value.

Parameters

rA	The LHS matrix

Returns: The diagonal max value

◆ GetColumn()

template<class TMatrixType , class TVectorType >

static void Kratos::TrilinosSpace< TMatrixType, TVectorType >::GetColumn	(	const unsigned int	j,
		const MatrixType &	rM,
		VectorType &	rX
	)

inlinestatic

Returns the column of the matrix in the given position.

rXi = rMij

Parameters

j	The position of the column
rM	The matrix considered
rX	The column considered

Todo:: Implement this method

◆ GetDiagonalNorm()

template<class TMatrixType , class TVectorType >

static double Kratos::TrilinosSpace< TMatrixType, TVectorType >::GetDiagonalNorm ( const MatrixType & rA )

inlinestatic

This method returns the diagonal norm considering for scaling the diagonal.

Parameters

rA	The LHS matrix

Returns: The diagonal norm

◆ GetMaxDiagonal()

template<class TMatrixType , class TVectorType >

static double Kratos::TrilinosSpace< TMatrixType, TVectorType >::GetMaxDiagonal ( const MatrixType & rA )

inlinestatic

This method returns the diagonal max value.

Parameters

rA	The LHS matrix

Returns: The diagonal max value

◆ GetMinDiagonal()

template<class TMatrixType , class TVectorType >

static double Kratos::TrilinosSpace< TMatrixType, TVectorType >::GetMinDiagonal ( const MatrixType & rA )

inlinestatic

This method returns the diagonal min value.

Parameters

rA	The LHS matrix

Returns: The diagonal min value

◆ GetScaleNorm()

template<class TMatrixType , class TVectorType >

static double Kratos::TrilinosSpace< TMatrixType, TVectorType >::GetScaleNorm	(	const ProcessInfo &	rProcessInfo,
		const MatrixType &	rA,
		const SCALING_DIAGONAL	ScalingDiagonal = `SCALING_DIAGONAL::NO_SCALING`
	)

inlinestatic

This method returns the scale norm considering for scaling the diagonal.

Parameters

rProcessInfo	The problem process info
rA	The LHS matrix
ScalingDiagonal	The type of caling diagonal considered

Returns: The scale norm

◆ GetValue()

template<class TMatrixType , class TVectorType >

static double Kratos::TrilinosSpace< TMatrixType, TVectorType >::GetValue	(	const VectorType &	rX,
		const std::size_t	I
	)

inlinestatic

This function returns a value from a given vector according to a given index.

Parameters

rX	The vector from which values are to be gathered
I	The index of the value to be gathered

Returns: The value of the vector corresponding to the index I

◆ Info()

template<class TMatrixType , class TVectorType >

virtual std::string Kratos::TrilinosSpace< TMatrixType, TVectorType >::Info ( ) const

inlinevirtual

Turn back information as a string.

Returns: Info as a string.

◆ InplaceMult()

template<class TMatrixType , class TVectorType >

static void Kratos::TrilinosSpace< TMatrixType, TVectorType >::InplaceMult	(	VectorType &	rX,
		const double	A
	)

inlinestatic

Returns the multiplication of a vector by a scalar.

y = A*x Checks if a multiplication is needed and tries to do otherwise

Parameters

rX	The vector considered
A	The scalar considered

◆ IsDistributed()

template<class TMatrixType , class TVectorType >

static constexpr bool Kratos::TrilinosSpace< TMatrixType, TVectorType >::IsDistributed ( )

inlinestaticconstexpr

This function returns if we are in a distributed system.

Returns: True if we are in a distributed system, false otherwise (always true in this case)

◆ IsDistributedSpace()

template<class TMatrixType , class TVectorType >

static constexpr bool Kratos::TrilinosSpace< TMatrixType, TVectorType >::IsDistributedSpace ( )

inlinestaticconstexpr

Check if the TrilinosSpace is distributed.

This static member function checks whether the TrilinosSpace is distributed or not.

Returns: True if the space is distributed, false otherwise.

◆ KRATOS_CLASS_POINTER_DEFINITION()

template<class TMatrixType , class TVectorType >

Kratos::TrilinosSpace< TMatrixType, TVectorType >::KRATOS_CLASS_POINTER_DEFINITION ( TrilinosSpace< TMatrixType, TVectorType > )

Pointer definition of TrilinosSpace.

◆ Max()

template<class TMatrixType , class TVectorType >

static double Kratos::TrilinosSpace< TMatrixType, TVectorType >::Max ( const VectorType & rX )

inlinestatic

Returns the maximum value of the vector rX.

Parameters

rX	The vector considered

Returns: The maximum value of the vector rX

◆ Min()

template<class TMatrixType , class TVectorType >

static double Kratos::TrilinosSpace< TMatrixType, TVectorType >::Min ( const VectorType & rX )

inlinestatic

Returns the minimum value of the vector rX.

Parameters

rX	The vector considered

Returns: The minimum value of the vector rX

◆ Mult() [1/2]

template<class TMatrixType , class TVectorType >

static void Kratos::TrilinosSpace< TMatrixType, TVectorType >::Mult	(	const MatrixType &	rA,
		const MatrixType &	rB,
		MatrixType &	rC,
		const bool	CallFillCompleteOnResult = `true`,
		const bool	KeepAllHardZeros = `false`
	)

inlinestatic

Returns the multiplication matrix-matrix.

C = A*B

Parameters

rA	The first matrix considered
rB	The second matrix considered
rC	The result of the multiplication
CallFillCompleteOnResult	Optional argument, defaults to true. Power users may specify this argument to be false if they DON'T want this function to call C.FillComplete. (It is often useful to allow this function to call C.FillComplete, in cases where one or both of the input matrices are rectangular and it is not trivial to know which maps to use for the domain- and range-maps.)
KeepAllHardZeros	Optional argument, defaults to false. If true, Multiply, keeps all entries in C corresponding to hard zeros. If false, the following happens by case: AB^T, A^TB^T - Does not store entries caused by hard zeros in C. A^TB (unoptimized) - Hard zeros are always stored (this option has no effect) AB, A^T*B (optimized) - Hard zeros in corresponding to hard zeros in A are not stored, There are certain cases involving reuse of C, where this can be useful.

◆ Mult() [2/2]

template<class TMatrixType , class TVectorType >

static void Kratos::TrilinosSpace< TMatrixType, TVectorType >::Mult	(	const MatrixType &	rA,
		const VectorType &	rX,
		VectorType &	rY
	)

inlinestatic

Returns the multiplication of a matrix by a vector.

y = A*x

Parameters

rA	The matrix considered
rX	The vector considered
rY	The result of the multiplication

◆ PrintData()

template<class TMatrixType , class TVectorType >

virtual void Kratos::TrilinosSpace< TMatrixType, TVectorType >::PrintData ( std::ostream & rOStream ) const

inlinevirtual

Print object's data.

Parameters

rOStream The output stream to print on.

◆ PrintInfo()

template<class TMatrixType , class TVectorType >

virtual void Kratos::TrilinosSpace< TMatrixType, TVectorType >::PrintInfo ( std::ostream & rOStream ) const

inlinevirtual

Print information about this object.

Parameters

rOStream The output stream to print on.

◆ ReadMatrixMarket()

template<class TMatrixType , class TVectorType >

MatrixPointerType Kratos::TrilinosSpace< TMatrixType, TVectorType >::ReadMatrixMarket	(	const std::string	FileName,
		Epetra_MpiComm &	rComm
	)

inline

Read a matrix from a MatrixMarket file.

Parameters

rFileName	The name of the file to read
rComm	The MPI communicator

Returns: The matrix read from the file

◆ ReadMatrixMarketVector()

template<class TMatrixType , class TVectorType >

VectorPointerType Kratos::TrilinosSpace< TMatrixType, TVectorType >::ReadMatrixMarketVector	(	const std::string &	rFileName,
		Epetra_MpiComm &	rComm,
		const int	N
	)

inline

Read a vector from a MatrixMarket file.

Parameters

rFileName	The name of the file to read
rComm	The MPI communicator
N	The size of the vector

◆ Resize() [1/3]

template<class TMatrixType , class TVectorType >

static void Kratos::TrilinosSpace< TMatrixType, TVectorType >::Resize	(	MatrixType &	rA,
		const SizeType	m,
		const SizeType	n
	)

inlinestatic

Resizes a matrix.

Parameters

rA	The matrix to be resized
m	The new number of rows
n	The new number of columns

◆ Resize() [2/3]

template<class TMatrixType , class TVectorType >

static void Kratos::TrilinosSpace< TMatrixType, TVectorType >::Resize	(	VectorPointerType &	pX,
		const SizeType	n
	)

inlinestatic

Resizes a vector.

Parameters

pA	The pointer to the vector to be resized
n	The new size

◆ Resize() [3/3]

template<class TMatrixType , class TVectorType >

static void Kratos::TrilinosSpace< TMatrixType, TVectorType >::Resize	(	VectorType &	rX,
		const SizeType	n
	)

inlinestatic

Resizes a vector.

Parameters

rX	The vector to be resized
n	The new size

◆ ScaleAndAdd() [1/2]

template<class TMatrixType , class TVectorType >

static void Kratos::TrilinosSpace< TMatrixType, TVectorType >::ScaleAndAdd	(	const double	A,
		const VectorType &	rX,
		const double	B,
		const VectorType &	rY,
		VectorType &	rZ
	)

inlinestatic

Returns the unaliased addition of two vectors by a scalar.

rZ = (A * rX) + (B * rY)

Parameters

A	The scalar considered
rX	The first vector considered
B	The scalar considered
rY	The second vector considered
rZ	The resulting vector considered

◆ ScaleAndAdd() [2/2]

template<class TMatrixType , class TVectorType >

static void Kratos::TrilinosSpace< TMatrixType, TVectorType >::ScaleAndAdd	(	const double	A,
		const VectorType &	rX,
		const double	B,
		VectorType &	rY
	)

inlinestatic

Returns the unaliased addition of two vectors by a scalar.

rY = (A * rX) + (B * rY)

Parameters

A	The scalar considered
rX	The first vector considered
B	The scalar considered
rY	The resulting vector considered

◆ Set()

template<class TMatrixType , class TVectorType >

static void Kratos::TrilinosSpace< TMatrixType, TVectorType >::Set	(	VectorType &	rX,
		const DataType	A
	)

inlinestatic

assigns a scalar to a vector

rX = A

Parameters

rX	The vector considered
A	The scalar considered

◆ SetToZero() [1/2]

template<class TMatrixType , class TVectorType >

static void Kratos::TrilinosSpace< TMatrixType, TVectorType >::SetToZero ( MatrixType & rA )

inlinestatic

Sets a matrix to zero.

Parameters

rX	The matrix to be set

◆ SetToZero() [2/2]

template<class TMatrixType , class TVectorType >

static void Kratos::TrilinosSpace< TMatrixType, TVectorType >::SetToZero ( VectorType & rX )

inlinestatic

Sets a vector to zero.

Parameters

rX	The vector to be set

◆ SetValue()

template<class TMatrixType , class TVectorType >

static void Kratos::TrilinosSpace< TMatrixType, TVectorType >::SetValue	(	VectorType &	rX,
		IndexType	i,
		const double	value
	)

inlinestatic

Sets a value in a vector.

Parameters

rX	The vector considered
i	The index of the value considered
value	The value considered

◆ Size()

template<class TMatrixType , class TVectorType >

static IndexType Kratos::TrilinosSpace< TMatrixType, TVectorType >::Size ( const VectorType & rV )

inlinestatic

Returns size of vector rV.

Parameters

rV	The vector considered

Returns: The size of the vector

◆ Size1()

template<class TMatrixType , class TVectorType >

static IndexType Kratos::TrilinosSpace< TMatrixType, TVectorType >::Size1 ( MatrixType const & rM )

inlinestatic

Returns number of rows of rM.

Parameters

rM	The matrix considered

Returns: The number of rows of rM

◆ Size2()

template<class TMatrixType , class TVectorType >

static IndexType Kratos::TrilinosSpace< TMatrixType, TVectorType >::Size2 ( MatrixType const & rM )

inlinestatic

Returns number of columns of rM.

Parameters

rM	The matrix considered

Returns: The number of columns of rM

◆ TransposeMult() [1/2]

template<class TMatrixType , class TVectorType >

static void Kratos::TrilinosSpace< TMatrixType, TVectorType >::TransposeMult	(	const MatrixType &	rA,
		const MatrixType &	rB,
		MatrixType &	rC,
		const std::pair< bool, bool >	TransposeFlag = `{false, false}`,
		const bool	CallFillCompleteOnResult = `true`,
		const bool	KeepAllHardZeros = `false`
	)

inlinestatic

Returns the transpose multiplication matrix-matrix.

C = A*B

Parameters

rA	The first matrix considered
rB	The second matrix considered
rC	The result of the multiplication
TransposeFlag	Flags to transpose the matrices
CallFillCompleteOnResult	Optional argument, defaults to true. Power users may specify this argument to be false if they DON'T want this function to call C.FillComplete. (It is often useful to allow this function to call C.FillComplete, in cases where one or both of the input matrices are rectangular and it is not trivial to know which maps to use for the domain- and range-maps.)
KeepAllHardZeros	Optional argument, defaults to false. If true, Multiply, keeps all entries in C corresponding to hard zeros. If false, the following happens by case: AB^T, A^TB^T - Does not store entries caused by hard zeros in C. A^TB (unoptimized) - Hard zeros are always stored (this option has no effect) AB, A^T*B (optimized) - Hard zeros in corresponding to hard zeros in A are not stored, There are certain cases involving reuse of C, where this can be useful.

◆ TransposeMult() [2/2]

template<class TMatrixType , class TVectorType >

static void Kratos::TrilinosSpace< TMatrixType, TVectorType >::TransposeMult	(	const MatrixType &	rA,
		const VectorType &	rX,
		VectorType &	rY
	)

inlinestatic

Returns the transpose multiplication of a matrix by a vector.

y = AT*x

Parameters

rA	The matrix considered
rX	The vector considered
rY	The result of the multiplication

◆ TwoNorm() [1/2]

template<class TMatrixType , class TVectorType >

static double Kratos::TrilinosSpace< TMatrixType, TVectorType >::TwoNorm ( const MatrixType & rA )

inlinestatic

Returns the Frobenius norm of the matrix rX.

||rA||2

Parameters

rA	The matrix considered

Returns: The Frobenius norm of the matrix rX

◆ TwoNorm() [2/2]

template<class TMatrixType , class TVectorType >

static double Kratos::TrilinosSpace< TMatrixType, TVectorType >::TwoNorm ( const VectorType & rX )

inlinestatic

Returns the norm of the vector rX.

||rX||2

Parameters

rX	The vector considered

Returns: The norm of the vector rX

◆ UnaliasedAdd()

template<class TMatrixType , class TVectorType >

static void Kratos::TrilinosSpace< TMatrixType, TVectorType >::UnaliasedAdd	(	VectorType &	rX,
		const double	A,
		const VectorType &	rY
	)

inlinestatic

Returns the unaliased addition of a vector by a scalar times a vector.

X += A*y; Checks if a multiplication is needed and tries to do otherwise

Note: ATTENTION it is assumed no aliasing between rX and rY

Parameters

rX	The resulting vector considered
A	The scalar considered
rY	The multiplied vector considered

◆ WriteMatrixMarketMatrix()

template<class TMatrixType , class TVectorType >

template<class TOtherMatrixType >

static bool Kratos::TrilinosSpace< TMatrixType, TVectorType >::WriteMatrixMarketMatrix	(	const char *	pFileName,
		const TOtherMatrixType &	rM,
		const bool	Symmetric
	)

inlinestatic

Writes a matrix to a file in MatrixMarket format.

Parameters

pFileName	The name of the file to be written
rM	The matrix to be written
Symmetric	If the matrix is symmetric

Returns: True if the file was successfully written, false otherwise

◆ WriteMatrixMarketVector()

template<class TMatrixType , class TVectorType >

template<class VectorType >

static bool Kratos::TrilinosSpace< TMatrixType, TVectorType >::WriteMatrixMarketVector	(	const char *	pFileName,
		const VectorType &	rV
	)

inlinestatic

Writes a vector to a file in MatrixMarket format.

Parameters

pFileName	The name of the file to be written
rV	The vector to be written

Returns: True if the file was successfully written, false otherwise

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

/home/runner/work/Documentation/Documentation/master/applications/TrilinosApplication/trilinos_space.h

Public Member Functions

Type Definitions

Operations

Input and output

Detailed Description

template<class TMatrixType, class TVectorType> class Kratos::TrilinosSpace< TMatrixType, TVectorType >

Member Typedef Documentation

◆ DataType

◆ DofUpdaterPointerType

◆ DofUpdaterType

◆ IndexType

◆ MatrixPointerType

◆ MatrixType

◆ SizeType

◆ VectorPointerType

◆ VectorType

Constructor & Destructor Documentation

◆ TrilinosSpace()

◆ ~TrilinosSpace()

Member Function Documentation

◆ AssembleLHS()

◆ AssembleRHS()

◆ Assign()

◆ BDBtProductOperation()

◆ BtDBProductOperation()

◆ CheckAndCorrectZeroDiagonalValues()

◆ Clear() [1/2]

◆ Clear() [2/2]

◆ CombineMatricesGraphs()

◆ Copy() [1/2]

◆ Copy() [2/2]

◆ CopyMatrixValues()

◆ CreateDofUpdater()

◆ CreateEmptyMatrixPointer() [1/2]

◆ CreateEmptyMatrixPointer() [2/2]

◆ CreateEmptyVectorPointer() [1/2]

◆ CreateEmptyVectorPointer() [2/2]

◆ Dot()

◆ GatherValues()

◆ GetAveragevalueDiagonal()

◆ GetColumn()

◆ GetDiagonalNorm()

◆ GetMaxDiagonal()

◆ GetMinDiagonal()

◆ GetScaleNorm()

◆ GetValue()

◆ Info()

◆ InplaceMult()

◆ IsDistributed()

◆ IsDistributedSpace()

◆ KRATOS_CLASS_POINTER_DEFINITION()

◆ Max()

◆ Min()

◆ Mult() [1/2]

◆ Mult() [2/2]

◆ PrintData()

◆ PrintInfo()

◆ ReadMatrixMarket()

◆ ReadMatrixMarketVector()

◆ Resize() [1/3]

◆ Resize() [2/3]

◆ Resize() [3/3]

◆ ScaleAndAdd() [1/2]

◆ ScaleAndAdd() [2/2]

◆ Set()

◆ SetToZero() [1/2]

◆ SetToZero() [2/2]

◆ SetValue()

◆ Size()

◆ Size1()

◆ Size2()

◆ TransposeMult() [1/2]

◆ TransposeMult() [2/2]

◆ TwoNorm() [1/2]

◆ TwoNorm() [2/2]

◆ UnaliasedAdd()

◆ WriteMatrixMarketMatrix()

◆ WriteMatrixMarketVector()

template<class TMatrixType, class TVectorType>
class Kratos::TrilinosSpace< TMatrixType, TVectorType >