Routines and structure to handle the MUMPS setup. The structure is used as a context structure of a cs_param_sles_t structure. More...

Data Structures
struct	cs_param_mumps_t
	Set of parameters to specify additional options to MUMPS For more advanced settings, one has to use the cs_user_sles_mumps_hook function. Please also refer to the MUMPS user guide for more details. More...

Enumerations
enum	cs_param_mumps_facto_type_t { CS_PARAM_MUMPS_FACTO_LU , CS_PARAM_MUMPS_FACTO_LDLT_SYM , CS_PARAM_MUMPS_FACTO_LDLT_SPD , CS_PARAM_MUMPS_N_FACTO_TYPES }
	type of factorization to consider when using the MUMPS solver to solve a linear system More...

enum	cs_param_mumps_analysis_algo_t { CS_PARAM_MUMPS_ANALYSIS_AMD , CS_PARAM_MUMPS_ANALYSIS_QAMD , CS_PARAM_MUMPS_ANALYSIS_PORD , CS_PARAM_MUMPS_ANALYSIS_SCOTCH , CS_PARAM_MUMPS_ANALYSIS_PTSCOTCH , CS_PARAM_MUMPS_ANALYSIS_METIS , CS_PARAM_MUMPS_ANALYSIS_PARMETIS , CS_PARAM_MUMPS_ANALYSIS_AUTO , CS_PARAM_MUMPS_N_ANALYSIS_ALGOS }
	Type of algorithm to consider when using the MUMPS solver to perform the analysis step (renumbering and graph manipulation). Please refer to the MUMPS user guide for more details about the following algorithms. AMD, QAMD and PORD are available with MUMPS without any prerequesite. More...

enum	cs_param_mumps_memory_usage_t { CS_PARAM_MUMPS_MEMORY_CONSTRAINED , CS_PARAM_MUMPS_MEMORY_AUTO , CS_PARAM_MUMPS_MEMORY_CPU_DRIVEN , CS_PARAM_MUMPS_N_MEMORY_USAGES }
	Strategy for the memory usage inside MUMPS. More...

Functions
cs_param_mumps_t *	cs_param_mumps_create (void)
	Create and initialize with the default settings a new structure storing a set of parameters used when calling MUMPS. More...

cs_param_mumps_t *	cs_param_mumps_copy (const cs_param_mumps_t *mumpsp)
	Copy into a new structure the given set of parameters used when calling MUMPS. More...

void	cs_param_mumps_log (const char name, const cs_param_mumps_t mumpsp)
	Log the structure storing the set of parameters used with MUMPS. More...

Detailed Description

Routines and structure to handle the MUMPS setup. The structure is used as a context structure of a cs_param_sles_t structure.

Enumeration Type Documentation

◆ cs_param_mumps_analysis_algo_t

enum cs_param_mumps_analysis_algo_t

Type of algorithm to consider when using the MUMPS solver to perform the analysis step (renumbering and graph manipulation). Please refer to the MUMPS user guide for more details about the following algorithms. AMD, QAMD and PORD are available with MUMPS without any prerequesite.

Enumerator
CS_PARAM_MUMPS_ANALYSIS_AMD	AMD is a sequential algorithm which is well-suited for 2D problem (for 3D problems it induces a higher memeory consumption).
CS_PARAM_MUMPS_ANALYSIS_QAMD	QAMD is a sequential algorithm which is well-suited for 2D problem (for 3D problems it induces a higher memeory consumption).
CS_PARAM_MUMPS_ANALYSIS_PORD	PORD is a sequential algorithm which is a good trade-off when MUMPS is installed with no prerequisite such as METIS or Scotch.
CS_PARAM_MUMPS_ANALYSIS_SCOTCH	SCOTCH is a sequential algorithm which delivers the very good performance with 3D meshes, generally, better than PORD and not as good as METIS
CS_PARAM_MUMPS_ANALYSIS_PTSCOTCH	PTSCOTCH is a parallel version of the sequential SCOTCH algorithm
CS_PARAM_MUMPS_ANALYSIS_METIS	METIS is a sequential algorithm which delivers the best performance in case of 2D meshes.
CS_PARAM_MUMPS_ANALYSIS_PARMETIS	PARMETIS is a parallel version of the sequential METIS algorithm
CS_PARAM_MUMPS_ANALYSIS_AUTO	MUMPS decides what is the best choice among available algorithms. This is the default choice.
CS_PARAM_MUMPS_N_ANALYSIS_ALGOS

◆ cs_param_mumps_facto_type_t

enum cs_param_mumps_facto_type_t

type of factorization to consider when using the MUMPS solver to solve a linear system

Enumerator
CS_PARAM_MUMPS_FACTO_LU	LU factorization is the most generic factorization available with MUMPS. It can handle general matrices (block and/or unsymmetric matrices)
CS_PARAM_MUMPS_FACTO_LDLT_SYM	This factorization is a Cholesky factorization (L.D.Lt) for general symmetric matrices.
CS_PARAM_MUMPS_FACTO_LDLT_SPD	This factorization is devoted to SPD matrices and corresponds to a Cholesky factorization. This is more specific and thus more efficient than CS_PARAM_MUMPS_FACTO_LDLT_SYM.
CS_PARAM_MUMPS_N_FACTO_TYPES

◆ cs_param_mumps_memory_usage_t

enum cs_param_mumps_memory_usage_t

Strategy for the memory usage inside MUMPS.

Enumerator
CS_PARAM_MUMPS_MEMORY_CONSTRAINED	Strategy aiming at limiting the memory usage
CS_PARAM_MUMPS_MEMORY_AUTO	Strategy relying on the default settings
CS_PARAM_MUMPS_MEMORY_CPU_DRIVEN	Strategy aiming at the best CPU time
CS_PARAM_MUMPS_N_MEMORY_USAGES

Function Documentation

◆ cs_param_mumps_copy()

cs_param_mumps_t * cs_param_mumps_copy ( const cs_param_mumps_t * mumpsp )

Copy into a new structure the given set of parameters used when calling MUMPS.

Parameters

[in] mumpsp set of mumps parameters

Returns: a pointer to a new allocated structure

◆ cs_param_mumps_create()

cs_param_mumps_t * cs_param_mumps_create ( void )

Create and initialize with the default settings a new structure storing a set of parameters used when calling MUMPS.

Returns: a pointer to a new allocated structure

◆ cs_param_mumps_log()

void cs_param_mumps_log	(	const char *	name,
		const cs_param_mumps_t *	mumpsp
	)

Log the structure storing the set of parameters used with MUMPS.