Converts the field type (eg. double to float) to benchmark single precision preconditioners. More...

#include <PreconditionerConvertFieldTypeAdapter.hpp>

Inheritance diagram for Opm::cuistl::PreconditionerConvertFieldTypeAdapter< CudaPreconditionerType, M, X, Y, l >:

Public Types
using	matrix_type = typename std::remove_const< M >::type
	The matrix type the preconditioner is for. More...

using	domain_type = X
	The domain type of the preconditioner. More...

using	range_type = Y
	The range type of the preconditioner. More...

using	field_type = typename X::field_type
	The field type of the preconditioner. More...

using	domain_type_to = typename CudaPreconditionerType::domain_type

using	range_type_to = typename CudaPreconditionerType::range_type
	The range type of the preconditioner. More...

using	field_type_to = typename domain_type_to::field_type
	The field type of the preconditioner. More...

using	block_type = typename domain_type::block_type

using	XTo = Dune::BlockVector< Dune::FieldVector< field_type_to, block_type::dimension > >

using	YTo = Dune::BlockVector< Dune::FieldVector< field_type_to, block_type::dimension > >

using	matrix_type_to = typename Dune::BCRSMatrix< Dune::FieldMatrix< field_type_to, block_type::dimension, block_type::dimension > >

Public Member Functions
	PreconditionerConvertFieldTypeAdapter (const M &matrix)
	Constructor. More...

virtual void	pre (X &x, Y &b) override
	Not used at the moment. More...

virtual void	apply (X &v, const Y &d) override
	Apply the preconditoner. More...

virtual void	post (X &x) override
	Not used at the moment. More...

virtual Dune::SolverCategory::Category	category () const override
	Category of the preconditioner (see SolverCategory::Category) More...

virtual void	update () override

const matrix_type_to &	getConvertedMatrix () const

void	setUnderlyingPreconditioner (const std::shared_ptr< CudaPreconditionerType > &conditioner)

Detailed Description

template<class CudaPreconditionerType, class M, class X, class Y, int l = 1>
class Opm::cuistl::PreconditionerConvertFieldTypeAdapter< CudaPreconditionerType, M, X, Y, l >

Converts the field type (eg. double to float) to benchmark single precision preconditioners.

Note: This is not a fast conversion, it is simply meant to benchmark the potential of some preconditioners on consumer grade GPUs where the double precision performance is often artificially limited.; In theory this can do any field_type conversion that is meaningful, but it is only tested on double to float conversion.; Remember to set the underlying preconditioner with setUnderlyingPreconditioner (should use the matrix from getConvertedMatrix()); One could simply change the constructor design by accepting a creator function for the underlying preconditioner. For the current use cases this is however not needed.

To use this, use something like the following code:

#include <opm/simulators/linalg/cuistl/PreconditionerConvertFieldTypeAdapter.hpp>
#include <opm/simulators/linalg/ParallelOverlappingILU0.hpp>
 
using XDouble = Dune::BlockVector<Dune::FieldVector<double, 2>>;
using MDouble = Dune::FieldMatrix<double, 2, 2>;
using SpMatrixDouble = Dune::BCRSMatrix<MDouble>;
using XFloat = Dune::BlockVector<Dune::FieldVector<float, 2>>;
using MFloat = Dune::FieldMatrix<float, 2, 2>;
using SpMatrixFloat = Dune::BCRSMatrix<MFloat>;
 
template<class ParallelInfo>
void applyILU0AsFloat(const MDouble& matrix, const XDouble& x, XDouble& y) {
 
    using FloatILU0 = typename Opm::ParallelOverlappingILU0<MFloat, XFloat, XFloat, ParallelInfo>;
    using DoubleToFloatConverter = typename Opm::cuistl::PreconditionerConvertFieldTypeAdapter<FloatILU0, MDouble,
    XDouble, XDouble>;
 
    // Note that we do not need to make a new instance for every invocation, this
    // is just done for this example
    auto doubleToFloatConverter = DoubleToFloatConverter(matrix);
    const auto& convertedMatrix = doubleToFloatConverter.getConvertedMatrix();
 
    auto floatILU0 = std::make_shared<FloatILU0>(convertedMatrix, 0, 1.0, 0);
 
    doubleToFloatConverter.setUnderlyingPreconditioner(floatILU0);
 
    // This will convert x and y to float, then call floatILU0.apply on the converted arguments
    doubleToFloatConverter.apply(x, y);
}

Member Typedef Documentation

◆ block_type

template<class CudaPreconditionerType , class M , class X , class Y , int l = 1>

using Opm::cuistl::PreconditionerConvertFieldTypeAdapter< CudaPreconditionerType, M, X, Y, l >::block_type = typename domain_type::block_type

◆ domain_type

template<class CudaPreconditionerType , class M , class X , class Y , int l = 1>

using Opm::cuistl::PreconditionerConvertFieldTypeAdapter< CudaPreconditionerType, M, X, Y, l >::domain_type = X

The domain type of the preconditioner.

◆ domain_type_to

template<class CudaPreconditionerType , class M , class X , class Y , int l = 1>

using Opm::cuistl::PreconditionerConvertFieldTypeAdapter< CudaPreconditionerType, M, X, Y, l >::domain_type_to = typename CudaPreconditionerType::domain_type

◆ field_type

template<class CudaPreconditionerType , class M , class X , class Y , int l = 1>

using Opm::cuistl::PreconditionerConvertFieldTypeAdapter< CudaPreconditionerType, M, X, Y, l >::field_type = typename X::field_type

The field type of the preconditioner.

◆ field_type_to

template<class CudaPreconditionerType , class M , class X , class Y , int l = 1>

using Opm::cuistl::PreconditionerConvertFieldTypeAdapter< CudaPreconditionerType, M, X, Y, l >::field_type_to = typename domain_type_to::field_type

The field type of the preconditioner.

◆ matrix_type

template<class CudaPreconditionerType , class M , class X , class Y , int l = 1>

using Opm::cuistl::PreconditionerConvertFieldTypeAdapter< CudaPreconditionerType, M, X, Y, l >::matrix_type = typename std::remove_const<M>::type

The matrix type the preconditioner is for.

◆ matrix_type_to

template<class CudaPreconditionerType , class M , class X , class Y , int l = 1>

using Opm::cuistl::PreconditionerConvertFieldTypeAdapter< CudaPreconditionerType, M, X, Y, l >::matrix_type_to = typename Dune::BCRSMatrix<Dune::FieldMatrix<field_type_to, block_type::dimension, block_type::dimension> >

◆ range_type

template<class CudaPreconditionerType , class M , class X , class Y , int l = 1>

using Opm::cuistl::PreconditionerConvertFieldTypeAdapter< CudaPreconditionerType, M, X, Y, l >::range_type = Y

The range type of the preconditioner.

◆ range_type_to

template<class CudaPreconditionerType , class M , class X , class Y , int l = 1>

using Opm::cuistl::PreconditionerConvertFieldTypeAdapter< CudaPreconditionerType, M, X, Y, l >::range_type_to = typename CudaPreconditionerType::range_type

The range type of the preconditioner.

◆ XTo

template<class CudaPreconditionerType , class M , class X , class Y , int l = 1>

using Opm::cuistl::PreconditionerConvertFieldTypeAdapter< CudaPreconditionerType, M, X, Y, l >::XTo = Dune::BlockVector<Dune::FieldVector<field_type_to, block_type::dimension> >

◆ YTo

template<class CudaPreconditionerType , class M , class X , class Y , int l = 1>

using Opm::cuistl::PreconditionerConvertFieldTypeAdapter< CudaPreconditionerType, M, X, Y, l >::YTo = Dune::BlockVector<Dune::FieldVector<field_type_to, block_type::dimension> >

Constructor & Destructor Documentation

◆ PreconditionerConvertFieldTypeAdapter()

template<class CudaPreconditionerType , class M , class X , class Y , int l = 1>

Opm::cuistl::PreconditionerConvertFieldTypeAdapter< CudaPreconditionerType, M, X, Y, l >::PreconditionerConvertFieldTypeAdapter ( const M & matrix )

inlineexplicit

Constructor.

Parameters

A	The matrix to operate on.

Note: After the PreconditionerConvertFieldTypeAdapter you can get the converted matrix by calling getConvertedMatrix(), which in turn can be used to create the underlying preconditioner. Once the underlying preconditioner has been called, this must be supplied to setUnderlyingPreconditioner.

Member Function Documentation

◆ apply()

template<class CudaPreconditionerType , class M , class X , class Y , int l = 1>

virtual void Opm::cuistl::PreconditionerConvertFieldTypeAdapter< CudaPreconditionerType, M, X, Y, l >::apply	(	X &	v,
		const Y &	d
	)

inlineoverridevirtual

Apply the preconditoner.

◆ category()

template<class CudaPreconditionerType , class M , class X , class Y , int l = 1>

virtual Dune::SolverCategory::Category Opm::cuistl::PreconditionerConvertFieldTypeAdapter< CudaPreconditionerType, M, X, Y, l >::category ( ) const

inlineoverridevirtual

Category of the preconditioner (see SolverCategory::Category)

◆ getConvertedMatrix()

template<class CudaPreconditionerType , class M , class X , class Y , int l = 1>

const matrix_type_to & Opm::cuistl::PreconditionerConvertFieldTypeAdapter< CudaPreconditionerType, M, X, Y, l >::getConvertedMatrix ( ) const

inline

◆ post()

template<class CudaPreconditionerType , class M , class X , class Y , int l = 1>

virtual void Opm::cuistl::PreconditionerConvertFieldTypeAdapter< CudaPreconditionerType, M, X, Y, l >::post ( X & x )

inlineoverridevirtual

Not used at the moment.

◆ pre()

template<class CudaPreconditionerType , class M , class X , class Y , int l = 1>

virtual void Opm::cuistl::PreconditionerConvertFieldTypeAdapter< CudaPreconditionerType, M, X, Y, l >::pre	(	X &	x,
		Y &	b
	)

inlineoverridevirtual

Not used at the moment.

◆ setUnderlyingPreconditioner()

template<class CudaPreconditionerType , class M , class X , class Y , int l = 1>

void Opm::cuistl::PreconditionerConvertFieldTypeAdapter< CudaPreconditionerType, M, X, Y, l >::setUnderlyingPreconditioner ( const std::shared_ptr< CudaPreconditionerType > & conditioner )

inline

◆ update()

template<class CudaPreconditionerType , class M , class X , class Y , int l = 1>

virtual void Opm::cuistl::PreconditionerConvertFieldTypeAdapter< CudaPreconditionerType, M, X, Y, l >::update ( )

inlineoverridevirtual

Implements Dune::PreconditionerWithUpdate< X, Y >.

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

PreconditionerConvertFieldTypeAdapter.hpp

Public Types

Public Member Functions

Detailed Description

Member Typedef Documentation

◆ block_type

◆ domain_type

◆ domain_type_to

◆ field_type

◆ field_type_to

◆ matrix_type

◆ matrix_type_to

◆ range_type

◆ range_type_to

◆ XTo

◆ YTo

Constructor & Destructor Documentation

◆ PreconditionerConvertFieldTypeAdapter()

Member Function Documentation

◆ apply()

◆ category()

◆ getConvertedMatrix()

◆ post()

◆ pre()

◆ setUnderlyingPreconditioner()

◆ update()