A strategy that takes a single inverse factory and uses that for all inverses. If no mass matrix is passed in the diagonal of the 1,1 block is used. More...

#include <Teko_InvLSCStrategy.hpp>

Inheritance diagram for Teko::NS::InvLSCStrategy:

[legend]

Public Member Functions
virtual void	setPressureStabMatrix (const Teko::LinearOp &psm)

virtual void	initializeState (const BlockedLinearOp &A, LSCPrecondState *state) const
	Initialize the state object using this blocked linear operator. More...

void	computeInverses (const BlockedLinearOp &A, LSCPrecondState *state) const

virtual void	setEigSolveParam (int sz)
	Set the number of power series iterations to use when finding the spectral radius. More...

virtual int	getEigSolveParam ()
	Return the number of power series iterations to use when finding the spectral radius. More...

virtual void	setUseFullLDU (bool val)
	Set to true to use the Full LDU decomposition, false otherwise. More...

virtual void	setRowZeroing (bool val)
	Set to true to zero the rows of F when computing the spectral radius. More...

virtual void	setMassMatrix (const LinearOp &mass)
	set the mass matrix to use in computing the scaling More...

virtual void	setHScaling (const LinearOp &hScaling)

virtual void	setHScaling (const MultiVector &hScaling)

virtual void	setWScaling (const MultiVector &wScaling)

Constructors
	InvLSCStrategy ()

	InvLSCStrategy (const Teuchos::RCP< InverseFactory > &factory, bool rzn=false)

	InvLSCStrategy (const Teuchos::RCP< InverseFactory > &factory, LinearOp &mass, bool rzn=false)

	InvLSCStrategy (const Teuchos::RCP< InverseFactory > &invFactF, const Teuchos::RCP< InverseFactory > &invFactS, bool rzn=false)

	InvLSCStrategy (const Teuchos::RCP< InverseFactory > &invFactF, const Teuchos::RCP< InverseFactory > &invFactS, LinearOp &mass, bool rzn=false)


virtual void	buildState (BlockedLinearOp &A, BlockPreconditionerState &state) const
	Functions inherited from LSCStrategy. More...

virtual LinearOp	getInvBQBt (const BlockedLinearOp &A, BlockPreconditionerState &state) const

virtual LinearOp	getInvBHBt (const BlockedLinearOp &A, BlockPreconditionerState &state) const

virtual LinearOp	getInvF (const BlockedLinearOp &A, BlockPreconditionerState &state) const

virtual LinearOp	getOuterStabilization (const BlockedLinearOp &A, BlockPreconditionerState &state) const

virtual LinearOp	getInnerStabilization (const BlockedLinearOp &, BlockPreconditionerState &) const

virtual LinearOp	getInvMass (const BlockedLinearOp &A, BlockPreconditionerState &state) const

virtual LinearOp	getHScaling (const BlockedLinearOp &A, BlockPreconditionerState &state) const

virtual bool	useFullLDU () const

virtual void	setSymmetric (bool isSymmetric)

virtual void	initializeFromParameterList (const Teuchos::ParameterList &pl, const InverseLibrary &invLib)
	Initialize from a parameter list. More...

virtual Teuchos::RCP < Teuchos::ParameterList >	getRequestedParameters () const
	For assiting in construction of the preconditioner. More...

virtual bool	updateRequestedParameters (const Teuchos::ParameterList &pl)
	For assiting in construction of the preconditioner. More...

Public Member Functions inherited from Teko::NS::LSCStrategy
void	setRequestHandler (const Teuchos::RCP< RequestHandler > &rh)
	This method sets the request handler for this object. More...

Teuchos::RCP< RequestHandler >	getRequestHandler () const
	This method gets the request handler uses by this object. More...

Detailed Description

A strategy that takes a single inverse factory and uses that for all inverses. If no mass matrix is passed in the diagonal of the 1,1 block is used.

A strategy that takes a single inverse factory and uses that for all inverses. Optionally the mass matrix can be passed in, if it is the diagonal is extracted and that is used to form the inverse approximation.

Definition at line 29 of file Teko_InvLSCStrategy.hpp.

Member Function Documentation

void Teko::NS::InvLSCStrategy::buildState	(	BlockedLinearOp &	A,
		BlockPreconditionerState &	state
	)		const

virtual

Functions inherited from LSCStrategy.

This informs the strategy object to build the state associated with this operator.

Parameters

[in]	A	The linear operator to be preconditioned by LSC.
[in]	state	State object for storying reusable information about the operator A.

Implements Teko::NS::LSCStrategy.

Definition at line 130 of file Teko_InvLSCStrategy.cpp.

LinearOp Teko::NS::InvLSCStrategy::getInvBQBt	(	const BlockedLinearOp &	A,
		BlockPreconditionerState &	state
	)		const

virtual

Get the inverse of $B Q_u^{-1} B^T$ .

Parameters

[in]	A	The linear operator to be preconditioned by LSC.
[in]	state	State object for storying reusable information about the operator A.

Returns: An (approximate) inverse of $B Q_u^{-1} B^T$ .

Implements Teko::NS::LSCStrategy.

Definition at line 165 of file Teko_InvLSCStrategy.cpp.

LinearOp Teko::NS::InvLSCStrategy::getInvBHBt	(	const BlockedLinearOp &	A,
		BlockPreconditionerState &	state
	)		const

virtual

Get the inverse of $B H B^T - \gamma C$ .

Parameters

[in]	A	The linear operator to be preconditioned by LSC.
[in]	state	State object for storying reusable information about the operator A.

Returns: An (approximate) inverse of $B H B^T - \gamma C$ .

Implements Teko::NS::LSCStrategy.

Definition at line 170 of file Teko_InvLSCStrategy.cpp.

LinearOp Teko::NS::InvLSCStrategy::getInvF	(	const BlockedLinearOp &	A,
		BlockPreconditionerState &	state
	)		const

virtual

Get the inverse of the $F$ block.

Parameters

[in]	A	The linear operator to be preconditioned by LSC.
[in]	state	State object for storying reusable information about the operator A.

Returns: An (approximate) inverse of .

Implements Teko::NS::LSCStrategy.

Definition at line 175 of file Teko_InvLSCStrategy.cpp.

LinearOp Teko::NS::InvLSCStrategy::getOuterStabilization	(	const BlockedLinearOp &	A,
		BlockPreconditionerState &	state
	)		const

virtual

Get the inverse for stabilizing the whole schur complement approximation.

Parameters

[in]	A	The linear operator to be preconditioned by LSC.
[in]	state	State object for storying reusable information about the operator A.

Returns: The operator to stabilize the whole Schur complement.

Implements Teko::NS::LSCStrategy.

Definition at line 180 of file Teko_InvLSCStrategy.cpp.

virtual LinearOp Teko::NS::InvLSCStrategy::getInnerStabilization	(	const BlockedLinearOp &	A,
		BlockPreconditionerState &	state
	)		const

inlinevirtual

Get the inverse to stablized stabilizing the Schur complement approximation using a placement on the ``inside''. That is what is the value for $C_I$ . This quantity may be null.

Parameters

[in]	A	The linear operator to be preconditioned by LSC.
[in]	state	State object for storying reusable information about the operator A.

Returns: The operator to stabilize the whole Schur complement.

Implements Teko::NS::LSCStrategy.

Definition at line 98 of file Teko_InvLSCStrategy.hpp.

LinearOp Teko::NS::InvLSCStrategy::getInvMass	(	const BlockedLinearOp &	A,
		BlockPreconditionerState &	state
	)		const

virtual

Get the inverse mass matrix.

Parameters

[in]	A	The linear operator to be preconditioned by LSC.
[in]	state	State object for storying reusable information about the operator A.

Returns: The inverse of the mass matrix .

Implements Teko::NS::LSCStrategy.

Definition at line 189 of file Teko_InvLSCStrategy.cpp.

LinearOp Teko::NS::InvLSCStrategy::getHScaling	(	const BlockedLinearOp &	A,
		BlockPreconditionerState &	state
	)		const

virtual

Get the $H$ scaling matrix.

Parameters

[in]	A	The linear operator to be preconditioned by LSC.
[in]	state	State object for storying reusable information about the operator A.

Returns: The scaling matrix.

Implements Teko::NS::LSCStrategy.

Definition at line 198 of file Teko_InvLSCStrategy.cpp.

virtual bool Teko::NS::InvLSCStrategy::useFullLDU ( ) const

inlinevirtual

Should the approximation of the inverse use a full LDU decomposition, or is a upper triangular approximation sufficient.

Returns: True if the full LDU decomposition should be used, otherwise only an upper triangular version is used.

Implements Teko::NS::LSCStrategy.

Definition at line 129 of file Teko_InvLSCStrategy.hpp.

virtual void Teko::NS::InvLSCStrategy::setSymmetric ( bool isSymmetric )

inlinevirtual

Tell strategy that this operator is supposed to be symmetric. Behavior of LSC is slightly different for non-symmetric case.

Parameters

[in] isSymmetric Is this operator symmetric?

Implements Teko::NS::LSCStrategy.

Definition at line 136 of file Teko_InvLSCStrategy.hpp.

void Teko::NS::InvLSCStrategy::initializeFromParameterList	(	const Teuchos::ParameterList &	pl,
		const InverseLibrary &	invLib
	)

virtual

Initialize from a parameter list.

Reimplemented from Teko::NS::LSCStrategy.

Definition at line 446 of file Teko_InvLSCStrategy.cpp.

Teuchos::RCP< Teuchos::ParameterList > Teko::NS::InvLSCStrategy::getRequestedParameters ( ) const

virtual

For assiting in construction of the preconditioner.

Reimplemented from Teko::NS::LSCStrategy.

Definition at line 517 of file Teko_InvLSCStrategy.cpp.

bool Teko::NS::InvLSCStrategy::updateRequestedParameters ( const Teuchos::ParameterList & pl )

virtual

For assiting in construction of the preconditioner.

Reimplemented from Teko::NS::LSCStrategy.

Definition at line 547 of file Teko_InvLSCStrategy.cpp.

virtual void Teko::NS::InvLSCStrategy::setPressureStabMatrix ( const Teko::LinearOp & psm )

inlinevirtual

When computing the Schur complement, use the passed in matrix instead of $C$ to stabilize the gradient operator.

Definition at line 151 of file Teko_InvLSCStrategy.hpp.

void Teko::NS::InvLSCStrategy::initializeState	(	const BlockedLinearOp &	A,
		LSCPrecondState *	state
	)		const

virtual

Initialize the state object using this blocked linear operator.

Definition at line 205 of file Teko_InvLSCStrategy.cpp.

void Teko::NS::InvLSCStrategy::computeInverses	(	const BlockedLinearOp &	A,
		LSCPrecondState *	state
	)		const

Compute the inverses required for the LSC Schur complement

Note: This method assumes that the BQBt and BHBt operators have been constructed.

Definition at line 385 of file Teko_InvLSCStrategy.cpp.

virtual void Teko::NS::InvLSCStrategy::setEigSolveParam ( int sz )

inlinevirtual

Set the number of power series iterations to use when finding the spectral radius.

Definition at line 167 of file Teko_InvLSCStrategy.hpp.

virtual int Teko::NS::InvLSCStrategy::getEigSolveParam ( )

inlinevirtual

Return the number of power series iterations to use when finding the spectral radius.

Definition at line 170 of file Teko_InvLSCStrategy.hpp.

virtual void Teko::NS::InvLSCStrategy::setUseFullLDU ( bool val )

inlinevirtual

Set to true to use the Full LDU decomposition, false otherwise.

Definition at line 173 of file Teko_InvLSCStrategy.hpp.

virtual void Teko::NS::InvLSCStrategy::setRowZeroing ( bool val )

inlinevirtual

Set to true to zero the rows of F when computing the spectral radius.

Definition at line 176 of file Teko_InvLSCStrategy.hpp.

virtual void Teko::NS::InvLSCStrategy::setMassMatrix ( const LinearOp & mass )

inlinevirtual

set the mass matrix to use in computing the scaling

Definition at line 179 of file Teko_InvLSCStrategy.hpp.

virtual void Teko::NS::InvLSCStrategy::setHScaling ( const LinearOp & hScaling )

inlinevirtual

Set the $H$ -Scaling operator used in $B H B^T$ . It is expected that this will be a diagonal matrix.

Definition at line 184 of file Teko_InvLSCStrategy.hpp.

virtual void Teko::NS::InvLSCStrategy::setHScaling ( const MultiVector & hScaling )

inlinevirtual

Set the $H$ -Scaling operator used in $B H B^T$ . This method takes a vector and constructs the diagonal matrix.

Definition at line 189 of file Teko_InvLSCStrategy.hpp.

virtual void Teko::NS::InvLSCStrategy::setWScaling ( const MultiVector & wScaling )

inlinevirtual

Set the $W$ -Scaling vector used in $B H B^T$ . This method takes a vector.

Definition at line 196 of file Teko_InvLSCStrategy.hpp.

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

Public Member Functions

Detailed Description

Member Function Documentation