IFPACK Development
Ifpack_ILUT Class Reference

Ifpack_ILUT: A class for constructing and using an incomplete LU factorization of a given Epetra_RowMatrix. More...

#include <Ifpack_ILUT.h>

Inheritance diagram for Ifpack_ILUT:
Inheritance graph
[legend]
Collaboration diagram for Ifpack_ILUT:
Collaboration graph
[legend]

List of all members.

Public Member Functions

 Ifpack_ILUT (const Epetra_RowMatrix *A)
 Ifpack_ILUT constuctor with variable number of indices per row.
virtual ~Ifpack_ILUT ()
 Ifpack_ILUT Destructor.
int SetParameters (Teuchos::ParameterList &parameterlis)
 Set parameters using a Teuchos::ParameterList object.
int Initialize ()
 Initialize L and U with values from user matrix A.
bool IsInitialized () const
 Returns true if the preconditioner has been successfully initialized.
int Compute ()
 Compute IC factor U using the specified graph, diagonal perturbation thresholds and relaxation parameters.
bool IsComputed () const
 If factor is completed, this query returns true, otherwise it returns false.
int ApplyInverse (const Epetra_MultiVector &X, Epetra_MultiVector &Y) const
 Returns the result of a Ifpack_ILUT forward/back solve on a Epetra_MultiVector X in Y.
int Apply (const Epetra_MultiVector &X, Epetra_MultiVector &Y) const
double Condest (const Ifpack_CondestType CT=Ifpack_Cheap, const int MaxIters=1550, const double Tol=1e-9, Epetra_RowMatrix *Matrix_in=0)
 Computed the estimated condition number and returns the value.
double Condest () const
 Returns the computed estimated condition number, or -1.0 if no computed.
int SetUseTranspose (bool UseTranspose_in)
 If set true, transpose of this operator will be applied.
double NormInf () const
 Returns 0.0 because this class cannot compute Inf-norm.
bool HasNormInf () const
 Returns false because this class cannot compute an Inf-norm.
bool UseTranspose () const
 Returns the current UseTranspose setting.
const Epetra_MapOperatorDomainMap () const
 Returns the Epetra_Map object associated with the domain of this operator.
const Epetra_MapOperatorRangeMap () const
 Returns the Epetra_Map object associated with the range of this operator.
const Epetra_CommComm () const
 Returns the Epetra_BlockMap object associated with the range of this matrix operator.
const Epetra_RowMatrixMatrix () const
 Returns a reference to the matrix to be preconditioned.
const Epetra_CrsMatrixL () const
 Returns a reference to the L factor.
const Epetra_CrsMatrixU () const
 Returns a reference to the U factor.
const char * Label () const
 Returns the label of this object.
int SetLabel (const char *Label_in)
 Sets the label for this object.
virtual ostream & Print (std::ostream &os) const
 Prints basic information on iostream. This function is used by operator<<.
virtual int NumInitialize () const
 Returns the number of calls to Initialize().
virtual int NumCompute () const
 Returns the number of calls to Compute().
virtual int NumApplyInverse () const
 Returns the number of calls to ApplyInverse().
virtual double InitializeTime () const
 Returns the time spent in Initialize().
virtual double ComputeTime () const
 Returns the time spent in Compute().
virtual double ApplyInverseTime () const
 Returns the time spent in ApplyInverse().
virtual double InitializeFlops () const
 Returns the number of flops in the initialization phase.
virtual double ComputeFlops () const
 Returns the number of flops in the computation phase.
virtual double ApplyInverseFlops () const
 Returns the number of flops in the application of the preconditioner.
double LevelOfFill () const
double RelaxValue () const
 Set relative threshold value.
double AbsoluteThreshold () const
 Get absolute threshold value.
double RelativeThreshold () const
 Get relative threshold value.
double DropTolerance () const
 Gets the dropping tolerance.
int NumGlobalNonzeros () const
 Returns the number of nonzero entries in the global graph.
int NumMyNonzeros () const
 Returns the number of nonzero entries in the local graph.

Detailed Description

Ifpack_ILUT: A class for constructing and using an incomplete LU factorization of a given Epetra_RowMatrix.

The Ifpack_ILUT class computes a "Relaxed" ILUT factorization with dual threshold dropping of small elements of a given Epetra_RowMatrix.

This implementation does not use the algorithm that is described in General description of incomplete factorizations. The algorithm drops entries in a row (i) of matrix A that are smaller than drop_tolerance even before the factorization of row i then computes the factorization for that row. This is different than the usual algorithm where the drop tolerance is applied to the factored rows.

The complete list of supported parameters is reported in page List of Supported Parameters.

Author:
Marzio Sala, SNL 9214.
Date:
Last modified on 22-Jan-05.

Definition at line 68 of file Ifpack_ILUT.h.


Member Function Documentation

int Ifpack_ILUT::ApplyInverse ( const Epetra_MultiVector X,
Epetra_MultiVector Y 
) const [virtual]

Returns the result of a Ifpack_ILUT forward/back solve on a Epetra_MultiVector X in Y.

Parameters:
X- (In) A Epetra_MultiVector of dimension NumVectors to solve for.
Y- (Out) A Epetra_MultiVector of dimension NumVectorscontaining result.
Returns:
Integer error code, set to 0 if successful.

Implements Ifpack_Preconditioner.

Definition at line 488 of file Ifpack_ILUT.cpp.

References Epetra_Time::ElapsedTime(), IsComputed(), Epetra_MultiVector::NumVectors(), Epetra_MultiVector::Pointers(), and Epetra_Time::ResetStartTime().

int Ifpack_ILUT::Compute ( ) [virtual]

Compute IC factor U using the specified graph, diagonal perturbation thresholds and relaxation parameters.

This function computes the RILU(k) factors L and U using the current:

  1. Ifpack_IlukGraph specifying the structure of L and U.
  2. Value for the RILU(k) relaxation parameter.
  3. Value for the a priori diagonal threshold values.

InitValues() must be called before the factorization can proceed.

Implements Ifpack_Preconditioner.

Definition at line 162 of file Ifpack_ILUT.cpp.

References AbsoluteThreshold(), Ifpack_HashTable::arrayify(), Comm(), Copy, DropTolerance(), Epetra_Time::ElapsedTime(), Epetra_RowMatrix::ExtractMyRowCopy(), Ifpack_HashTable::get(), Ifpack_HashTable::getNumEntries(), Initialize(), IsInitialized(), Epetra_RowMatrix::MaxNumEntries(), Epetra_RowMatrix::Multiply(), Epetra_RowMatrix::NumGlobalNonzeros(), Epetra_RowMatrix::NumMyRows(), Epetra_Comm::NumProc(), Epetra_MultiVector::Random(), RelativeThreshold(), RelaxValue(), Ifpack_HashTable::reset(), Epetra_Time::ResetStartTime(), Epetra_RowMatrix::RowMatrixRowMap(), Ifpack_HashTable::set(), and Epetra_Comm::SumAll().

int Ifpack_ILUT::Initialize ( ) [virtual]

Initialize L and U with values from user matrix A.

Copies values from the user's matrix into the nonzero pattern of L and U.

Parameters:
InA - User matrix to be factored.
Warning:
The graph of A must be identical to the graph passed in to Ifpack_IlukGraph constructor.

Implements Ifpack_Preconditioner.

Definition at line 127 of file Ifpack_ILUT.cpp.

References Comm(), Epetra_Time::ElapsedTime(), Matrix(), Epetra_RowMatrix::NumMyRows(), and Epetra_Time::ResetStartTime().

Referenced by Compute().

int Ifpack_ILUT::SetUseTranspose ( bool  UseTranspose_in) [inline, virtual]

If set true, transpose of this operator will be applied.

This flag allows the transpose of the given operator to be used implicitly. Setting this flag affects only the Apply() and ApplyInverse() methods. If the implementation of this interface does not support transpose use, this method should return a value of -1.

Parameters:
UseTranspose_in- (In) If true, multiply by the transpose of operator, otherwise just use operator.
Returns:
Always returns 0.

Implements Epetra_Operator.

Definition at line 156 of file Ifpack_ILUT.h.


The documentation for this class was generated from the following files:
 All Classes Files Functions Variables Enumerations Friends