.rst .pdf

Sparse BLAS Routines

The Intel® oneAPI Math Kernel Library provides a Data Parallel C++ interface to some of the Sparse BLAS routines. This section describes the Sparse BLAS routines included in the Intel® oneAPI Math Kernel Library (oneMKL).

Sparse BLAS provides basic operations on sparse vectors and matrices and separates them into three stages: initialization, analysis (also called inspector stage or optimize stage) and execution. For a given matrix, the analysis would typically be called one time and the execution may be called multiple times. During the analysis stage, the API inspects the matrix properties including size, sparsity pattern and available parallelism and can apply matrix format or structure changes to enable a more optimized algorithm. In the execution stage, multiple routine calls can take advantage of the analysis stage data in order to improve performance.

The supported sparse formats in Sparse BLAS routines are described in Sparse Storage Formats.

Routine or Function Group	Data Types	Description
sparse::init_matrix_handle	N/A	Initialize a sparse::matrix_handle_t object
sparse::release_matrix_handle	N/A	Release a sparse::matrix_handle_t object
sparse::set_csr_data	float, double	Set internal representation of sparse::matrix_handle_t to compressed sparse row (CSR) format with user provided CSR data arrays.
sparse::optimize_gemv	N/A	Perform internal optimizations for the sparse::gemv operation.
sparse::optimize_trmv	N/A	Perform internal optimizations for the sparse::trmv operation.
sparse::optimize_trsv	N/A	Perform internal optimizations for the sparse::trsv operation.
sparse::gemv	float, double	General sparse matrix-dense vector product
sparse::gemvdot	float, double	General sparse matrix-dense vector product with fused dot product
sparse::symv	float, double	Symmetric sparse matrix-dense vector product
sparse::trmv	float, double	Triangular sparse matrix-dense vector product
sparse::trsv	float, double	Triangular solve of sparse matrix against a dense vector.
sparse::gemm	float, double	General sparse matrix-dense matrix product
sparse::init_matmat_descr	N/A	Initialize the matmat_descr_t object. For use with sparse::matmat
sparse::set_matmat_data	N/A	Populate the matmat_descr_t object with desired operation description. For use with sparse::matmat
sparse::get_matmat_data	N/A	Query the operation description housed in the matmat_descr_t object. For use with sparse::matmat
sparse::release_matmat_descr	N/A	Release the matmat_descr_t object. For use with sparse::matmat
sparse::matmat	float, double	General sparse matrix-sparse matrix product

The currently supported sparse storage formats can be found in Sparse Storage Formats.

omatadd_batch oneapi::mkl::sparse::init_matrix_handle