#include <deal.II/lac/trilinos_solver.h>

Inheritance diagram for TrilinosWrappers::SolverBase:

Classes
struct	AdditionalData

Public Types
enum	SolverName { cg, cgs, gmres, bicgstab, tfqmr }

Public Member Functions
	SolverBase (SolverControl &cn, const AdditionalData &data=AdditionalData())

	SolverBase (const enum SolverName solver_name, SolverControl &cn, const AdditionalData &data=AdditionalData())

virtual	~SolverBase ()=default

void	solve (const SparseMatrix &A, MPI::Vector &x, const MPI::Vector &b, const PreconditionBase &preconditioner)

void	solve (const Epetra_Operator &A, MPI::Vector &x, const MPI::Vector &b, const PreconditionBase &preconditioner)

void	solve (const Epetra_Operator &A, MPI::Vector &x, const MPI::Vector &b, const Epetra_Operator &preconditioner)

void	solve (const Epetra_Operator &A, Epetra_MultiVector &x, const Epetra_MultiVector &b, const PreconditionBase &preconditioner)

void	solve (const Epetra_Operator &A, Epetra_MultiVector &x, const Epetra_MultiVector &b, const Epetra_Operator &preconditioner)

void	solve (const SparseMatrix &A, ::Vector< double > &x, const ::Vector< double > &b, const PreconditionBase &preconditioner)

void	solve (Epetra_Operator &A, ::Vector< double > &x, const ::Vector< double > &b, const PreconditionBase &preconditioner)

void	solve (const SparseMatrix &A, ::LinearAlgebra::distributed::Vector< double > &x, const ::LinearAlgebra::distributed::Vector< double > &b, const PreconditionBase &preconditioner)

void	solve (Epetra_Operator &A, ::LinearAlgebra::distributed::Vector< double > &x, const ::LinearAlgebra::distributed::Vector< double > &b, const PreconditionBase &preconditioner)

SolverControl &	control () const

template<>
void	set_preconditioner (AztecOO &solver, const PreconditionBase &preconditioner)

template<>
void	set_preconditioner (AztecOO &solver, const Epetra_Operator &preconditioner)

Static Public Member Functions
static ::ExceptionBase &	ExcTrilinosError (int arg1)

Public Attributes
enum TrilinosWrappers::SolverBase::SolverName	solver_name

Protected Attributes
SolverControl &	solver_control

Private Member Functions
template<typename Preconditioner >
void	do_solve (const Preconditioner &preconditioner)

template<typename Preconditioner >
void	set_preconditioner (AztecOO &solver, const Preconditioner &preconditioner)

Private Attributes
std::unique_ptr< Epetra_LinearProblem >	linear_problem

std::unique_ptr< AztecOO_StatusTest >	status_test

AztecOO	solver

const AdditionalData	additional_data

Detailed Description

Base class for solver classes using the Trilinos solvers. Since solvers in Trilinos are selected based on flags passed to a generic solver object, basically all the actual solver calls happen in this class, and derived classes simply set the right flags to select one solver or another, or to set certain parameters for individual solvers. For a general discussion on the Trilinos solver package AztecOO, we refer to the AztecOO user guide.

This solver class can also be used as a standalone class, where the respective Krylov method is set via the flag solver_name. This can be done at runtime (e.g., when parsing the solver from a ParameterList) and is similar to the deal.II class SolverSelector.

Author: Martin Kronbichler, 2008, 2009; extension for full compatibility with LinearOperator class: Jean-Paul Pelteret, 2015

Definition at line 68 of file trilinos_solver.h.

Member Enumeration Documentation

◆ SolverName

enum TrilinosWrappers::SolverBase::SolverName

Enumeration object that is set in the constructor of the derived classes and tells Trilinos which solver to use. This option can also be set in the user program, so one might use this base class instead of one of the specialized derived classes when the solver should be set at runtime. Currently enabled options are:

Enumerator
cg	Use the conjugate gradient (CG) algorithm.
cgs	Use the conjugate gradient squared (CGS) algorithm.
gmres	Use the generalized minimum residual (GMRES) algorithm.
bicgstab	Use the biconjugate gradient stabilized (BICGStab) algorithm.
tfqmr	Use the transpose-free quasi-minimal residual (TFQMR) method.

Definition at line 78 of file trilinos_solver.h.

Constructor & Destructor Documentation

◆ SolverBase() [1/2]

SolverBase< VectorType >::SolverBase	(	SolverControl &	cn,
		const AdditionalData &	data = `AdditionalData()`
	)

Constructor. Takes the solver control object and creates the solver.

Definition at line 49 of file trilinos_solver.cc.

◆ SolverBase() [2/2]

SolverBase< VectorType >::SolverBase	(	const enum SolverName	solver_name,
		SolverControl &	cn,
		const AdditionalData &	data = `AdditionalData()`
	)

Second constructor. This constructor takes an enum object that specifies the solver name and sets the appropriate Krylov method.

Definition at line 57 of file trilinos_solver.cc.

◆ ~SolverBase()

virtual TrilinosWrappers::SolverBase::~SolverBase ( )

virtualdefault

Destructor.

Member Function Documentation

◆ solve() [1/9]

void SolverBase< VectorType >::solve	(	const SparseMatrix &	A,
		MPI::Vector &	x,
		const MPI::Vector &	b,
		const PreconditionBase &	preconditioner
	)

Solve the linear system Ax=b. Depending on the information provided by derived classes and the object passed as a preconditioner, one of the linear solvers and preconditioners of Trilinos is chosen.

Definition at line 76 of file trilinos_solver.cc.

◆ solve() [2/9]

void SolverBase< VectorType >::solve	(	const Epetra_Operator &	A,
		MPI::Vector &	x,
		const MPI::Vector &	b,
		const PreconditionBase &	preconditioner
	)

Solve the linear system Ax=b where A is an operator. This function can be used for matrix free computation. Depending on the information provided by derived classes and the object passed as a preconditioner, one of the linear solvers and preconditioners of Trilinos is chosen.

Definition at line 96 of file trilinos_solver.cc.

◆ solve() [3/9]

void SolverBase< VectorType >::solve	(	const Epetra_Operator &	A,
		MPI::Vector &	x,
		const MPI::Vector &	b,
		const Epetra_Operator &	preconditioner
	)

Solve the linear system Ax=b where both A and its preconditioner are an operator. This function can be used when both A and the preconditioner are LinearOperators derived from a TrilinosPayload. Depending on the information provided by derived classes and the object passed as a preconditioner, one of the linear solvers and preconditioners of Trilinos is chosen.

Definition at line 116 of file trilinos_solver.cc.

◆ solve() [4/9]

void SolverBase< VectorType >::solve	(	const Epetra_Operator &	A,
		Epetra_MultiVector &	x,
		const Epetra_MultiVector &	b,
		const PreconditionBase &	preconditioner
	)

Solve the linear system Ax=b where A is an operator, and the vectors x and b are native Trilinos vector types. This function can be used when A is a LinearOperators derived from a TrilinosPayload. Depending on the information provided by derived classes and the object passed as a preconditioner, one of the linear solvers and preconditioners of Trilinos is chosen.

Definition at line 136 of file trilinos_solver.cc.

◆ solve() [5/9]

void SolverBase< VectorType >::solve	(	const Epetra_Operator &	A,
		Epetra_MultiVector &	x,
		const Epetra_MultiVector &	b,
		const Epetra_Operator &	preconditioner
	)

Solve the linear system Ax=b where both A and its preconditioner are an operator, and the vectors x and b are native Trilinos vector types. This function can be used when both A and the preconditioner are LinearOperators derived from a TrilinosPayload. Depending on the information provided by derived classes and the object passed as a preconditioner, one of the linear solvers and preconditioners of Trilinos is chosen.

Definition at line 156 of file trilinos_solver.cc.

◆ solve() [6/9]

void SolverBase< VectorType >::solve	(	const SparseMatrix &	A,
		::Vector< double > &	x,
		const ::Vector< double > &	b,
		const PreconditionBase &	preconditioner
	)

Solve the linear system Ax=b. Depending on the information provided by derived classes and the object passed as a preconditioner, one of the linear solvers and preconditioners of Trilinos is chosen. This class works with matrices according to the TrilinosWrappers format, but can take deal.II vectors as argument. Since deal.II are serial vectors (not distributed), this function does only what you expect in case the matrix is locally owned. Otherwise, an exception will be thrown.

Definition at line 174 of file trilinos_solver.cc.

◆ solve() [7/9]

void SolverBase< VectorType >::solve	(	Epetra_Operator &	A,
		::Vector< double > &	x,
		const ::Vector< double > &	b,
		const PreconditionBase &	preconditioner
	)

Solve the linear system Ax=b where A is an operator. This function can be used for matrix free computations. Depending on the information provided by derived classes and the object passed as a preconditioner, one of the linear solvers and preconditioners of Trilinos is chosen. This class works with matrices according to the TrilinosWrappers format, but can take deal.II vectors as argument. Since deal.II are serial vectors (not distributed), this function does only what you expect in case the matrix is locally owned. Otherwise, an exception will be thrown.

Definition at line 204 of file trilinos_solver.cc.

◆ solve() [8/9]

void SolverBase< VectorType >::solve	(	const SparseMatrix &	A,
		::LinearAlgebra::distributed::Vector< double > &	x,
		const ::LinearAlgebra::distributed::Vector< double > &	b,
		const PreconditionBase &	preconditioner
	)

Solve the linear system Ax=b for deal.II's parallel distributed vectors. Depending on the information provided by derived classes and the object passed as a preconditioner, one of the linear solvers and preconditioners of Trilinos is chosen.

Definition at line 225 of file trilinos_solver.cc.

◆ solve() [9/9]

void SolverBase< VectorType >::solve	(	Epetra_Operator &	A,
		::LinearAlgebra::distributed::Vector< double > &	x,
		const ::LinearAlgebra::distributed::Vector< double > &	b,
		const PreconditionBase &	preconditioner
	)

Solve the linear system Ax=b where A is an operator. This function can be used for matrix free computation. Depending on the information provided by derived classes and the object passed as a preconditioner, one of the linear solvers and preconditioners of Trilinos is chosen.

Definition at line 253 of file trilinos_solver.cc.

◆ control()

SolverControl & SolverBase< VectorType >::control ( ) const

Access to object that controls convergence.

Definition at line 68 of file trilinos_solver.cc.

◆ do_solve()

template<typename Preconditioner >

template void SolverBase< VectorType >::do_solve ( const Preconditioner & preconditioner )

private

The solve function is used to set properly the Epetra_LinearProblem, once it is done this function solves the linear problem.

Definition at line 414 of file trilinos_solver.cc.

◆ set_preconditioner() [1/3]

template<typename Preconditioner >

void TrilinosWrappers::SolverBase::set_preconditioner	(	AztecOO &	solver,
		const Preconditioner &	preconditioner
	)

private

A function that sets the preconditioner that the solver will apply

◆ set_preconditioner() [2/3]

template<>

void SolverBase::set_preconditioner	(	AztecOO &	solver,
		const PreconditionBase &	preconditioner
	)

Definition at line 565 of file trilinos_solver.cc.

◆ set_preconditioner() [3/3]

template<>

void SolverBase::set_preconditioner	(	AztecOO &	solver,
		const Epetra_Operator &	preconditioner
	)

Definition at line 583 of file trilinos_solver.cc.

Member Data Documentation

◆ solver_name

enum TrilinosWrappers::SolverBase::SolverName TrilinosWrappers::SolverBase::solver_name

◆ solver_control

SolverControl& TrilinosWrappers::SolverBase::solver_control

protected

Reference to the object that controls convergence of the iterative solver. In fact, for these Trilinos wrappers, Trilinos does so itself, but we copy the data from this object before starting the solution process, and copy the data back into it afterwards.

Definition at line 302 of file trilinos_solver.h.

◆ linear_problem

std::unique_ptr<Epetra_LinearProblem> TrilinosWrappers::SolverBase::linear_problem

private

A structure that collects the Trilinos sparse matrix, the right hand side vector and the solution vector, which is passed down to the Trilinos solver.

Definition at line 325 of file trilinos_solver.h.

◆ status_test

std::unique_ptr<AztecOO_StatusTest> TrilinosWrappers::SolverBase::status_test

private

A structure that contains a Trilinos object that can query the linear solver and determine whether the convergence criterion have been met.

Definition at line 331 of file trilinos_solver.h.

◆ solver

AztecOO TrilinosWrappers::SolverBase::solver

private

A structure that contains the Trilinos solver and preconditioner objects.

Definition at line 337 of file trilinos_solver.h.

◆ additional_data

const AdditionalData TrilinosWrappers::SolverBase::additional_data

private

Store a copy of the flags for this particular solver.

Definition at line 342 of file trilinos_solver.h.

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

include/deal.II/lac/trilinos_solver.h
source/lac/trilinos_solver.cc

Classes

Public Types

Public Member Functions

Static Public Member Functions

Public Attributes

Protected Attributes

Private Member Functions

Private Attributes

Detailed Description

Member Enumeration Documentation

◆ SolverName

Constructor & Destructor Documentation

◆ SolverBase() [1/2]

◆ SolverBase() [2/2]

◆ ~SolverBase()

Member Function Documentation

◆ solve() [1/9]

◆ solve() [2/9]

◆ solve() [3/9]

◆ solve() [4/9]

◆ solve() [5/9]

◆ solve() [6/9]

◆ solve() [7/9]

◆ solve() [8/9]

◆ solve() [9/9]

◆ control()

◆ do_solve()

◆ set_preconditioner() [1/3]

◆ set_preconditioner() [2/3]

◆ set_preconditioner() [3/3]

Member Data Documentation

◆ solver_name

◆ solver_control

◆ linear_problem

◆ status_test

◆ solver

◆ additional_data