LinearAlgebra Namespace Reference

Namespaces
namespace	CUDAWrappers
namespace	distributed
namespace	EpetraWrappers
namespace	TpetraWrappers

Classes
class	BlockVector
class	ReadWriteVector
class	Vector
class	VectorSpaceVector

Typedefs
using	CommunicationPatternBase = Utilities::MPI::CommunicationPatternBase

Enumerations
enum class	OrthogonalizationStrategy { modified_gram_schmidt , classical_gram_schmidt }

Functions
template<typename Number >
void	set_zero_mean_value (VectorSpaceVector< Number > &vector)
	TEMPL_COPY_CONSTRUCTOR (double, float)
	TEMPL_COPY_CONSTRUCTOR (float, double)
	TEMPL_COPY_CONSTRUCTOR (std::complex< double >, std::complex< float >)
	TEMPL_COPY_CONSTRUCTOR (std::complex< float >, std::complex< double >)

Detailed Description

A namespace for vector classes.

This namespace contains various classes that provide wrappers to vector classes from different external libraries like Trilinos (EPetra) or PETSc and native implementations like LinearAlgebra::distributed::Vector.

The different vector classes are derived from VectorSpaceVector to provide a joint interface for vector space operations, are derived from ReadWriteVector (or ReadWriteVector itself), or both. The separation of vector space operations (like norms or vector additions) through VectorSpaceVector and element access through ReadWriteVector are by design and improve performance.

Typedef Documentation

CommunicationPatternBase

using LinearAlgebra::CommunicationPatternBase = typedef Utilities::MPI::CommunicationPatternBase

Alias for Utilities::MPI::CommunicationPatternBase. This class was originally defined in the LinearAlgebra namespace but is now used for more general purposes.

Definition at line 33 of file communication_pattern_base.h.

Enumeration Type Documentation

OrthogonalizationStrategy

enum class LinearAlgebra::OrthogonalizationStrategy

strong

Supported orthogonalization strategies within SolverGMRES and SolverFGMRES.

Enumerator
modified_gram_schmidt	Use modified Gram-Schmidt algorithm.
classical_gram_schmidt	Use classical Gram-Schmidt algorithm. Since this approach works on multi-vectors with a single global reduction (of multiple elements), it is more efficient than the modified Gram-Schmidt algorithm. However, it is less stable in terms of roundoff error propagation, requiring additional re-orthogonalization steps more frequently.

Definition at line 30 of file orthogonalization.h.

Function Documentation

set_zero_mean_value()

template<typename Number >

void LinearAlgebra::set_zero_mean_value

(

VectorSpaceVector< Number > &

vector

)

Shift all entries of the vector by a constant factor so that the mean value of the vector becomes zero.

Definition at line 296 of file vector_space_vector.h.

TEMPL_COPY_CONSTRUCTOR() [1/4]

LinearAlgebra::TEMPL_COPY_CONSTRUCTOR	(	double	,
		float
	)

TEMPL_COPY_CONSTRUCTOR() [2/4]

LinearAlgebra::TEMPL_COPY_CONSTRUCTOR	(	float	,
		double
	)

TEMPL_COPY_CONSTRUCTOR() [3/4]

LinearAlgebra::TEMPL_COPY_CONSTRUCTOR	(	std::complex< double >	,
		std::complex< float >
	)

TEMPL_COPY_CONSTRUCTOR() [4/4]

LinearAlgebra::TEMPL_COPY_CONSTRUCTOR	(	std::complex< float >	,
		std::complex< double >
	)