A module for wrapping a Distributed Sparse Matrix. More...

Data Types
type	distributedsparsematrix_wrp
	A wrapper for the sparse matrix data type. More...

Functions/Subroutines
pure subroutine, public	constructemptydistributedsparsematrix_wrp (ih_this, matrix_dim)
	Wrap the constructor of an empty sparse, distributed, matrix. More...

subroutine, public	constructfrommatrixmarket_wrp (ih_this, file_name, name_size)
	Construct distributed sparse matrix from a matrix market file in parallel. More...

subroutine, public	constructfrombinary_wrp (ih_this, file_name, name_size)
	Construct a distributed sparse matrix from a binary file in parallel. More...

pure subroutine, public	copydistributedsparsematrix_wrp (ih_matA, ih_matB)
	Copy a distributed sparse matrix in a safe way. More...

pure subroutine, public	destructdistributedsparsematrix_wrp (ih_this)
	Destruct a distributed sparse matrix. More...

subroutine, public	writetobinary_wrp (ih_this, file_name, name_size)
	Save a distributed sparse matrix to a file. More...

subroutine, public	writetomatrixmarket_wrp (ih_this, file_name, name_size)
	Save a distributed sparse matrix to a matrix market file. More...

subroutine, public	fillfromtripletlist_wrp (ih_this, ih_triplet_list)
	This routine fills in a matrix based on local triplet lists. More...

pure subroutine, public	filldistributedidentity_wrp (ih_this)
	Fill in the values of a distributed matrix with the identity matrix. More...

subroutine, public	filldistributedpermutation_wrp (ih_this, ih_permutation, permute_rows)
	Fill in the values of a distributed matrix with a permutation matrix. More...

pure subroutine, public	getactualdimension_wrp (ih_this, mat_dimension)
	Wrap the Actual Dimension accessor. More...

pure subroutine, public	getlogicaldimension_wrp (ih_this, mat_dimension)
	Wrap the Logical Dimension accessor. More...

pure subroutine, public	gettripletlist_wrp (ih_this, ih_triplet_list)
	Extracts a triplet list of the data that is stored on this process. More...

subroutine, public	incrementdistributedsparsematrix_wrp (ih_matA, ih_matB, alpha_in, threshold_in)
	Matrix B = alpha*Matrix A + Matrix B (AXPY) More...

real(ntreal) function, public	dotdistributedsparsematrix_wrp (ih_matA, ih_matB)
	product = dot(matA,matB) More...

subroutine, public	distributedpairwisemultiply_wrp (ih_matA, ih_matB, ih_matC)
	Pairwise multiply two matrices together. More...

subroutine, public	distributedgemm_wrp (ih_matA, ih_matB, ih_matC, alpha_in, beta_in, threshold_in, ih_memory_pool_in)
	Multiply two matrices together, and add to the third. More...

pure subroutine, public	scaledistributedsparsematrix_wrp (ih_this, constant)
	Will scale a distributed sparse matrix by a constant. More...

real(ntreal) function, public	distributedsparsenorm_wrp (ih_this)
	Compute the norm of a distributed sparse matrix along the rows. More...

real(ntreal) function, public	trace_wrp (ih_this)
	Compute the trace of the matrix. More...

Detailed Description

A module for wrapping a Distributed Sparse Matrix.

Function/Subroutine Documentation

◆ constructemptydistributedsparsematrix_wrp()

pure subroutine, public distributedsparsematrixmodule_wrp::constructemptydistributedsparsematrix_wrp	(	integer(kind=c_int), dimension(size_wrp), intent(inout)	ih_this,
		integer(kind=c_int), intent(in)	matrix_dim
	)

Wrap the constructor of an empty sparse, distributed, matrix.

Parameters

[in,out]	ih_this	the matrix to be constructed.
[in]	matrix_dim	the dimension of the full matrix.

◆ constructfrombinary_wrp()

subroutine, public distributedsparsematrixmodule_wrp::constructfrombinary_wrp	(	integer(kind=c_int), dimension(size_wrp), intent(inout)	ih_this,
		character(kind=c_char), dimension(name_size), intent(in)	file_name,
		integer(kind=c_int), intent(in)	name_size
	)

Construct a distributed sparse matrix from a binary file in parallel.

Parameters

[out]	ih_this	the file being constructed.
[in]	file_name	name of the file to read.
[in]	name_size	the number of characters in the file name.

◆ constructfrommatrixmarket_wrp()

subroutine, public distributedsparsematrixmodule_wrp::constructfrommatrixmarket_wrp	(	integer(kind=c_int), dimension(size_wrp), intent(inout)	ih_this,
		character(kind=c_char), dimension(name_size), intent(in)	file_name,
		integer(kind=c_int), intent(in)	name_size
	)

Construct distributed sparse matrix from a matrix market file in parallel.

Parameters

[out]	ih_this	the file being constructed.
[in]	file_name	name of the file to read.
[in]	name_size	the number of characters in the file name.

◆ copydistributedsparsematrix_wrp()

pure subroutine, public distributedsparsematrixmodule_wrp::copydistributedsparsematrix_wrp	(	integer(kind=c_int), dimension(size_wrp), intent(in)	ih_matA,
		integer(kind=c_int), dimension(size_wrp), intent(inout)	ih_matB
	)

Copy a distributed sparse matrix in a safe way.

Parameters

[in]	ih_matA	matrix to copy
[in,out]	ih_matB	= matA

◆ destructdistributedsparsematrix_wrp()

pure subroutine, public distributedsparsematrixmodule_wrp::destructdistributedsparsematrix_wrp ( integer(kind=c_int), dimension(size_wrp), intent(inout) ih_this )

Destruct a distributed sparse matrix.

Parameters

[in,out] ih_this the matrix to destruct

◆ distributedgemm_wrp()

subroutine, public distributedsparsematrixmodule_wrp::distributedgemm_wrp	(	integer(kind=c_int), dimension(size_wrp), intent(in)	ih_matA,
		integer(kind=c_int), dimension(size_wrp), intent(in)	ih_matB,
		integer(kind=c_int), dimension(size_wrp), intent(inout)	ih_matC,
		real(ntreal), intent(in)	alpha_in,
		real(ntreal), intent(in)	beta_in,
		real(ntreal), intent(in)	threshold_in,
		integer(kind=c_int), dimension(size_wrp), intent(inout)	ih_memory_pool_in
	)

Multiply two matrices together, and add to the third.

Parameters

[in]	ih_matA	Matrix A
[in]	ih_matB	Matrix B
[out]	ih_matC	= alphamatAmatB + beta*matC
[in]	alpha_in	scales the multiplication
[in]	beta_in	scales matrix we sum on to
[in]	threshold_in	for flushing values to zero.
[in,out]	ih_memory_pool_in	a memory pool that can be used for the calculation.

◆ distributedpairwisemultiply_wrp()

subroutine, public distributedsparsematrixmodule_wrp::distributedpairwisemultiply_wrp	(	integer(kind=c_int), dimension(size_wrp), intent(in)	ih_matA,
		integer(kind=c_int), dimension(size_wrp), intent(in)	ih_matB,
		integer(kind=c_int), dimension(size_wrp), intent(inout)	ih_matC
	)

Pairwise multiply two matrices together.

Parameters

[in]	ih_matA	Matrix A
[in]	ih_matB	Matrix B
[out]	ih_matC	= matA*matB

◆ distributedsparsenorm_wrp()

real(ntreal) function, public distributedsparsematrixmodule_wrp::distributedsparsenorm_wrp ( integer(kind=c_int), dimension(size_wrp), intent(in) ih_this )

Compute the norm of a distributed sparse matrix along the rows.

Parameters

[in] ih_this the matrix to compute the norm of.

Returns: the norm value of the full distributed sparse matrix.

◆ dotdistributedsparsematrix_wrp()

real(ntreal) function, public distributedsparsematrixmodule_wrp::dotdistributedsparsematrix_wrp	(	integer(kind=c_int), dimension(size_wrp), intent(in)	ih_matA,
		integer(kind=c_int), dimension(size_wrp), intent(inout)	ih_matB
	)

product = dot(matA,matB)

Parameters

[in]	ih_matA	Matrix A
[in]	ih_matB	Matrix B

Returns: product dot product of the two matrices

◆ filldistributedidentity_wrp()

pure subroutine, public distributedsparsematrixmodule_wrp::filldistributedidentity_wrp ( integer(kind=c_int), dimension(size_wrp), intent(inout) ih_this )

Fill in the values of a distributed matrix with the identity matrix.

Parameters

[in,out] ih_this the matrix being filled.

◆ filldistributedpermutation_wrp()

subroutine, public distributedsparsematrixmodule_wrp::filldistributedpermutation_wrp	(	integer(kind=c_int), dimension(size_wrp), intent(inout)	ih_this,
		integer(kind=c_int), dimension(size_wrp), intent(in)	ih_permutation,
		logical(kind=c_bool), intent(in)	permute_rows
	)

Fill in the values of a distributed matrix with a permutation matrix.

Parameters

[in,out]	ih_this	the matrix being filled.
[in]	ih_permutation	the perumutation to fill with.
[in]	permute_rows	whether to permute rows or columns.

◆ fillfromtripletlist_wrp()

subroutine, public distributedsparsematrixmodule_wrp::fillfromtripletlist_wrp	(	integer(kind=c_int), dimension(size_wrp), intent(inout)	ih_this,
		integer(kind=c_int), dimension(size_wrp), intent(in)	ih_triplet_list
	)

This routine fills in a matrix based on local triplet lists.

Parameters

[in,out]	ih_this	the matrix being filled.
[in]	ih_triplet_list	the triplet list of values.

◆ getactualdimension_wrp()

pure subroutine, public distributedsparsematrixmodule_wrp::getactualdimension_wrp	(	integer(kind=c_int), dimension(size_wrp), intent(in)	ih_this,
		integer(kind=c_int), intent(out)	mat_dimension
	)

Wrap the Actual Dimension accessor.

Parameters

[in]	ih_this	handle to the matrix.
[out]	mat_dimension	the size of the matrix.

◆ getlogicaldimension_wrp()

pure subroutine, public distributedsparsematrixmodule_wrp::getlogicaldimension_wrp	(	integer(kind=c_int), dimension(size_wrp), intent(in)	ih_this,
		integer(kind=c_int), intent(out)	mat_dimension
	)

Wrap the Logical Dimension accessor.

Parameters

[in]	ih_this	handle to the matrix.
[out]	mat_dimension	the size of the matrix.

◆ gettripletlist_wrp()

pure subroutine, public distributedsparsematrixmodule_wrp::gettripletlist_wrp	(	integer(kind=c_int), dimension(size_wrp), intent(in)	ih_this,
		integer(kind=c_int), dimension(size_wrp), intent(inout)	ih_triplet_list
	)

Extracts a triplet list of the data that is stored on this process.

Parameters

[in]	ih_this	the distributed sparse matrix.
[in,out]	ih_triplet_list	the list to fill.

◆ incrementdistributedsparsematrix_wrp()

subroutine, public distributedsparsematrixmodule_wrp::incrementdistributedsparsematrix_wrp	(	integer(kind=c_int), dimension(size_wrp), intent(in)	ih_matA,
		integer(kind=c_int), dimension(size_wrp), intent(inout)	ih_matB,
		real(ntreal), intent(in)	alpha_in,
		real(ntreal), intent(in)	threshold_in
	)

Matrix B = alpha*Matrix A + Matrix B (AXPY)

Parameters

[in]	ih_matA	Matrix A
[in,out]	ih_matB	Matrix B
[in]	alpha_in	multiplier.
[in]	threshold_in	for flushing values to zero.

◆ scaledistributedsparsematrix_wrp()

pure subroutine, public distributedsparsematrixmodule_wrp::scaledistributedsparsematrix_wrp	(	integer(kind=c_int), dimension(size_wrp), intent(inout)	ih_this,
		real(ntreal), intent(in)	constant
	)

Will scale a distributed sparse matrix by a constant.

Parameters

[in,out]	ih_this	Matrix to scale.
[in]	constant	scale factor.

◆ trace_wrp()

real(ntreal) function, public distributedsparsematrixmodule_wrp::trace_wrp ( integer(kind=c_int), dimension(size_wrp), intent(in) ih_this )

Compute the trace of the matrix.

Parameters

[in] ih_this the matrix to compute the norm of.

Returns: the trace value of the full distributed sparse matrix.

◆ writetobinary_wrp()

subroutine, public distributedsparsematrixmodule_wrp::writetobinary_wrp	(	integer(kind=c_int), dimension(size_wrp), intent(in)	ih_this,
		character(kind=c_char), dimension(name_size), intent(in)	file_name,
		integer(kind=c_int), intent(in)	name_size
	)

Save a distributed sparse matrix to a file.

Parameters

[in]	ih_this	the Matrix to write.
[in]	file_name	name of the file to write to.
[in]	name_size	the number of characters in the file name.

◆ writetomatrixmarket_wrp()

subroutine, public distributedsparsematrixmodule_wrp::writetomatrixmarket_wrp	(	integer(kind=c_int), dimension(size_wrp), intent(in)	ih_this,
		character(kind=c_char), dimension(name_size), intent(in)	file_name,
		integer(kind=c_int), intent(in)	name_size
	)

Save a distributed sparse matrix to a matrix market file.

Parameters

[in]	ih_this	the Matrix to write.
[in]	file_name	name of the file to write to.
[in]	name_size	the number of characters in the file name.

Data Types

Functions/Subroutines

Detailed Description

Function/Subroutine Documentation

◆ constructemptydistributedsparsematrix_wrp()

◆ constructfrombinary_wrp()

◆ constructfrommatrixmarket_wrp()

◆ copydistributedsparsematrix_wrp()

◆ destructdistributedsparsematrix_wrp()

◆ distributedgemm_wrp()

◆ distributedpairwisemultiply_wrp()

◆ distributedsparsenorm_wrp()

◆ dotdistributedsparsematrix_wrp()

◆ filldistributedidentity_wrp()

◆ filldistributedpermutation_wrp()

◆ fillfromtripletlist_wrp()

◆ getactualdimension_wrp()

◆ getlogicaldimension_wrp()

◆ gettripletlist_wrp()

◆ incrementdistributedsparsematrix_wrp()

◆ scaledistributedsparsematrix_wrp()

◆ trace_wrp()

◆ writetobinary_wrp()

◆ writetomatrixmarket_wrp()