The incremental compressed row storage sparse matrix data structure. More...

#include <ICRS.hpp>

Inheritance diagram for ICRS< T, _i_value >:

Collaboration diagram for ICRS< T, _i_value >:

Public Member Functions
	ICRS ()
	Base constructor. More...

	ICRS (std::string file, T zero=0)
	Base constructor. More...

	ICRS (const ULI number_of_nonzeros, const ULI number_of_rows, const ULI number_of_cols, T zero)
	Base constructor which only initialises the internal arrays. More...

	ICRS (ICRS< T > &toCopy)
	Copy constructor. More...

	ICRS (std::vector< Triplet< T > > &input, const ULI m, const ULI n, const T zero=0)
	Constructor which transforms a collection of input triplets to CRS format. More...

virtual void	load (std::vector< Triplet< T > > &input, const ULI m, const ULI n, const T zero)

virtual void	getFirstIndexPair (ULI &row, ULI &col)
	Returns the first nonzero index, per reference. More...

void	getStartingPos (ULI &row_start, ULI &column_start)
	Gets starting position (first nonzero location)

void	setStartingPos (const ULI row_start, const ULI column_start)
	Sets starting position of matrix multiplication. More...

virtual void	zxa (const T __restrict__ pDataX, T __restrict__ pDataZ)
	In-place z=xA function. More...

virtual void	zax (const T __restrict__ pDataX, T __restrict__ pDataZ)
	In-place z=Ax function. More...

template<size_t k>
void	ZaX (const T __restrict__ const __restrict__ const X, T __restrict__ const __restrict__ const Z)

template<size_t k>
void	ZXa (const T __restrict__ const __restrict__ const X, T __restrict__ const __restrict__ const Z)

	~ICRS ()
	Base deconstructor. More...

virtual size_t	bytesUsed ()
	Function to query the amount of storage required by this sparse matrix. More...

Public Member Functions inherited from SparseMatrix< T, ULI >
	SparseMatrix ()
	Base constructor. More...

	SparseMatrix (const ULInzs, const ULInr, const ULInc, const T zero)
	Base constructor. More...

virtual	~SparseMatrix ()
	Base deconstructor. More...

void	loadFromFile (const std::string file, const T zero=0)
	Function which loads a matrix from a matrix market file. More...

virtual unsigned long int	m ()
	Queries the number of rows this matrix contains. More...

virtual unsigned long int	n ()
	Queries the number of columns this matrix contains. More...

virtual unsigned long int	nzs ()
	Queries the number of nonzeroes stored in this matrix. More...

virtual T *	mv (const T *x)
	Calculates and returns z=Ax. More...

Public Member Functions inherited from Matrix< T >
	Matrix ()
	Base constructor. More...

virtual	~Matrix ()
	Base deconstructor. More...

virtual void	zax (const T __restrict__ x, T __restrict__ z, const size_t k, const clockid_t clock_id=0, double *elapsed_time=NULL)
	Wrapper function to call the zax kernel multiple times successively, while timing the duration of the operation. More...

template<size_t k>
void	ZaX (const T __restrict__ const __restrict__ const X, T __restrict__ const __restrict__ const Z)
	In-place Z=AX function, where A is m x n, Z = m x k, and X is n x k. More...

template<size_t k>
void	ZXa (const T __restrict__ const __restrict__ const X, T __restrict__ const __restrict__ const Z)
	In-place Z=XA function, where A is m x n, Z = k x n, and X is k x m. More...

virtual void	zxa (const T __restrict__ x, T __restrict__ z, const unsigned long int repeat, const clockid_t clock_id=0, double *elapsed_time=NULL)
	Wrapper function to call the zxa kernel multiple times successively, while timing the operation duration. More...

Static Public Member Functions
static int	compareTriplets (const void left, const void right)
	Comparison function used for sorting input data. More...

Static Public Attributes
static const size_t	fillIn = 0
	Fill-in field for interoperability with vecBICRS. More...

Protected Attributes
T *	ds
	Array containing the actual nnz non-zeros. More...

ULI	r_start
	Start position, row.

ULI	c_start
	Start position, column.

_i_value *	c_ind
	Array containing the column jumps. More...

_i_value *	r_ind
	Array containing the row jumps. More...

size_t	bytes
	Remembers the number of bytes allocated. More...

Protected Attributes inherited from SparseMatrix< T, ULI >
ULI	nor
	Number of rows. More...

ULI	noc
	Number of columns.

ULI	nnz
	Number of non-zeros. More...

Additional Inherited Members
Public Attributes inherited from SparseMatrix< T, ULI >
T	zero_element
	The element considered to be zero. More...

Detailed Description

template<typename T, typename _i_value = ULI>
class ICRS< T, _i_value >

The incremental compressed row storage sparse matrix data structure.

Constructor & Destructor Documentation

template<typename T, typename _i_value = ULI>

ICRS< T, _i_value >::ICRS ( )

inline

Base constructor.

template<typename T, typename _i_value = ULI>

ICRS< T, _i_value >::ICRS	(	std::string	file,
		T	zero = `0`
	)

inline

Base constructor.

Will read in from Matrix Market file.

See also: SparseMatrix::SparseMatrix( file, zero )

References SparseMatrix< T, ULI >::loadFromFile().

template<typename T, typename _i_value = ULI>

ICRS< T, _i_value >::ICRS	(	const ULI	number_of_nonzeros,
		const ULI	number_of_rows,
		const ULI	number_of_cols,
		T	zero
	)

inline

Base constructor which only initialises the internal arrays.

Note that to gain a valid ICRS structure, these arrays have to be filled by some external mechanism (i.e., after calling this constructor, the internal arrays contain garbage, resuling in invalid datastructure).

Parameters

number_of_nonzeros	The number of non-zeros to be stored.
number_of_rows	The number of rows of the matrix to be stored.
number_of_cols	The number of columns of the matrix to be stored.
zero	Which element is considered to be the zero element.

References ICRS< T, _i_value >::bytes, ICRS< T, _i_value >::c_ind, ICRS< T, _i_value >::ds, SparseMatrix< T, ULI >::nnz, and ICRS< T, _i_value >::r_ind.

template<typename T, typename _i_value = ULI>

ICRS< T, _i_value >::ICRS ( ICRS< T > & toCopy )

inline

Copy constructor.

Parameters

toCopy Reference to the ICRS datastructure to copy.

References ICRS< T, _i_value >::bytes, ICRS< T, _i_value >::c_ind, ICRS< T, _i_value >::c_start, ICRS< T, _i_value >::ds, SparseMatrix< T, ULI >::nnz, SparseMatrix< T, ULI >::noc, SparseMatrix< T, ULI >::nor, ICRS< T, _i_value >::r_ind, ICRS< T, _i_value >::r_start, and SparseMatrix< T, ULI >::zero_element.

template<typename T, typename _i_value = ULI>

ICRS< T, _i_value >::ICRS	(	std::vector< Triplet< T > > &	input,
		const ULI	m,
		const ULI	n,
		const T	zero = `0`
	)

inline

Constructor which transforms a collection of input triplets to CRS format.

The input collection is considered to have at most one triplet with unique pairs of indices. Unspecified behaviour occurs when this assumption is not met.

Parameters

input	The input collection of triplets (i,j,val).
m	The number of rows of the input matrix.
n	The number of columns of the input matrix.
zero	Which element is considered zero.

References ICRS< T, _i_value >::load().

template<typename T, typename _i_value = ULI>

ICRS< T, _i_value >::~ICRS ( )

inline

Base deconstructor.

References ICRS< T, _i_value >::c_ind, ICRS< T, _i_value >::ds, and ICRS< T, _i_value >::r_ind.

Member Function Documentation

template<typename T, typename _i_value = ULI>

virtual size_t ICRS< T, _i_value >::bytesUsed ( )

inlinevirtual

Function to query the amount of storage required by this sparse matrix.

Returns: The size of the sparse matrix in bytes.

Implements Matrix< T >.

References ICRS< T, _i_value >::bytes.

template<typename T, typename _i_value = ULI>

static int ICRS< T, _i_value >::compareTriplets	(	const void *	left,
		const void *	right
	)

inlinestatic

Comparison function used for sorting input data.

References Triplet< T >::i(), and Triplet< T >::j().

Referenced by ICRS< T, _i_value >::load().

template<typename T, typename _i_value = ULI>

virtual void ICRS< T, _i_value >::getFirstIndexPair	(	ULI &	row,
		ULI &	col
	)

inlinevirtual

Returns the first nonzero index, per reference.

Implements SparseMatrix< T, ULI >.

References ICRS< T, _i_value >::c_start, and ICRS< T, _i_value >::r_start.

template<typename T, typename _i_value = ULI>

virtual void ICRS< T, _i_value >::load	(	std::vector< Triplet< T > > &	input,
		const ULI	m,
		const ULI	n,
		const T	zero
	)

inlinevirtual

See also: SparseMatrix::load

Implements SparseMatrix< T, ULI >.

References ICRS< T, _i_value >::bytes, ICRS< T, _i_value >::c_ind, ICRS< T, _i_value >::c_start, ICRS< T, _i_value >::compareTriplets(), ICRS< T, _i_value >::ds, Triplet< T >::i(), Triplet< T >::j(), SparseMatrix< T, ULI >::m(), SparseMatrix< T, ULI >::n(), SparseMatrix< T, ULI >::nnz, SparseMatrix< T, ULI >::noc, SparseMatrix< T, ULI >::nor, ICRS< T, _i_value >::r_ind, ICRS< T, _i_value >::r_start, Triplet< T >::value, and SparseMatrix< T, ULI >::zero_element.

Referenced by ICRS< T, _i_value >::ICRS().

template<typename T, typename _i_value = ULI>

void ICRS< T, _i_value >::setStartingPos	(	const ULI	row_start,
		const ULI	column_start
	)

inline

Sets starting position of matrix multiplication.

(Useful for example when the input/output vectors will be shifted before passed on to this class' zax method.)

Parameters

row_start	New row start location
column_start	New column start location

References ICRS< T, _i_value >::c_start, and ICRS< T, _i_value >::r_start.

template<typename T, typename _i_value = ULI>

virtual void ICRS< T, _i_value >::zax	(	const T *__restrict__	pDataX,
		T *__restrict__	pDataZ
	)

inlinevirtual

In-place z=Ax function.

Adapted from the master thesis of Joris Koster, Utrecht University.

Parameters

pDataX	Pointer to array x to multiply by the current matrix (Ax).
pDataZ	Pointer to result array. Must be pre-allocated and its elements set to zero for correct results.

Implements SparseMatrix< T, ULI >.

References ICRS< T, _i_value >::c_ind, ICRS< T, _i_value >::c_start, ICRS< T, _i_value >::ds, SparseMatrix< T, ULI >::nnz, SparseMatrix< T, ULI >::noc, SparseMatrix< T, ULI >::nor, ICRS< T, _i_value >::r_ind, and ICRS< T, _i_value >::r_start.

template<typename T, typename _i_value = ULI>

template<size_t k>

void ICRS< T, _i_value >::ZaX	(	const T __restrict__ const __restrict__ const	X,
		T __restrict__ const __restrict__ const	Z
	)

inline

See also: Matrix::ZaX

References ICRS< T, _i_value >::c_ind, ICRS< T, _i_value >::c_start, ICRS< T, _i_value >::ds, ICRS< T, _i_value >::r_ind, and ICRS< T, _i_value >::r_start.

template<typename T, typename _i_value = ULI>

virtual void ICRS< T, _i_value >::zxa	(	const T *__restrict__	pDataX,
		T *__restrict__	pDataZ
	)

inlinevirtual

In-place z=xA function.

Adapted from the master thesis of Joris Koster, Utrecht University.

Parameters

pDataX	Pointer to array x to multiply by the current matrix (Ax).
pDataZ	Pointer to result array. Must be pre-allocated and its elements set to zero for correct results.

Implements SparseMatrix< T, ULI >.

References ICRS< T, _i_value >::c_ind, ICRS< T, _i_value >::c_start, ICRS< T, _i_value >::ds, SparseMatrix< T, ULI >::nnz, SparseMatrix< T, ULI >::noc, SparseMatrix< T, ULI >::nor, ICRS< T, _i_value >::r_ind, and ICRS< T, _i_value >::r_start.

template<typename T, typename _i_value = ULI>

template<size_t k>