Inheritance diagram for MEDCoupling::MPIProcessorGroup:

Collaboration diagram for MEDCoupling::MPIProcessorGroup:

Public Member Functions
bool	containsMyRank () const

ProcessorGroup *	createComplementProcGroup () const

ProcessorGroup *	createProcGroup () const

virtual MPIProcessorGroup *	deepCopy () const

virtual ProcessorGroup *	fuse (const ProcessorGroup &) const

const MPI_Comm *	getComm () const

MPI_Comm	getWorldComm ()

void	intersect (ProcessorGroup &)

	MPIProcessorGroup (const CommInterface &interface)

	MPIProcessorGroup (const CommInterface &interface, int pstart, int pend, const MPI_Comm &world_comm=MPI_COMM_WORLD)

	MPIProcessorGroup (const CommInterface &interface, std::map< std::string, std::set< int >> proc_ids_by_name, const std::string &simCodeTag, const MPI_Comm &world_comm=MPI_COMM_WORLD)

	MPIProcessorGroup (const CommInterface &interface, std::set< int > proc_ids, const MPI_Comm &world_comm=MPI_COMM_WORLD)

	MPIProcessorGroup (const MPIProcessorGroup &other)

	MPIProcessorGroup (const ProcessorGroup &proc_group, std::set< int > proc_ids)

int	myRank () const

void	release ()

int	translateRank (const ProcessorGroup *group, int rank) const

virtual	~MPIProcessorGroup ()

Public Member Functions inherited from MEDCoupling::ProcessorGroup
bool	contains (int rank) const

const CommInterface &	getCommInterface () const

virtual const std::set< int > &	getProcIDs () const

virtual const std::set< int > &	getProcIDsByName (const std::string &simCodeTag) const

	ProcessorGroup (const CommInterface &interface)

	ProcessorGroup (const CommInterface &interface, int start, int end)

	ProcessorGroup (const CommInterface &interface, std::map< std::string, std::set< int >> proc_ids_by_name, const std::string &simCodeTag)

	ProcessorGroup (const CommInterface &interface, std::set< int > proc_ids)

	ProcessorGroup (const ProcessorGroup &other)

	ProcessorGroup (const ProcessorGroup &proc_group, std::set< int > proc_ids)

int	size () const

virtual	~ProcessorGroup ()

Additional Inherited Members
Protected Attributes inherited from MEDCoupling::ProcessorGroup
const CommInterface	_comm_interface

std::set< int >	_proc_ids

std::map< std::string, std::set< int > >	_proc_ids_by_name

Detailed Description

The MPIProcessorGroup class represents a set of distinct "processors" (computation nodes) in a MPI code. It is used to define the MPI topology of code couplings.

Groups can be set up in various ways, the most common being the use of the MPIProcessorGroup(Comminterface, int pfirst, int plast) constructor.

The following code excerpt creates two processor groups on respectively 3 and 2 processors.

int main()
{
MPI_Init(&argc,&argv);
CommInterface comm_interface;
MPIProcessorGroup codeA_group(comm_interface, 0, 2);  // groups processors 0, 1 and 2
MPIProcessorGroup codeB_group(comm_interface, 3, 4);  // groups processors 3 and 4

...
}

Constructor & Destructor Documentation

◆ MPIProcessorGroup() [1/6]

MEDCoupling::MPIProcessorGroup::MPIProcessorGroup ( const CommInterface & interface )

Creates a processor group that is based on all the processors of MPI_COMM_WORLD .This routine must be called by all processors in MPI_COMM_WORLD.

Parameters

interface CommInterface object giving access to the MPI communication layer

References MEDCoupling::ProcessorGroup::_comm_interface, MEDCoupling::ProcessorGroup::_proc_ids, MEDCoupling::CommInterface::commGroup(), MEDCoupling::CommInterface::commSize(), and MEDCoupling::ProcessorGroup::size().

Referenced by createComplementProcGroup(), createProcGroup(), deepCopy(), and fuse().

◆ MPIProcessorGroup() [2/6]

MEDCoupling::MPIProcessorGroup::MPIProcessorGroup	(	const CommInterface &	interface,
		std::set< int >	proc_ids,
		const MPI_Comm &	world_comm = `MPI_COMM_WORLD`
	)

Creates a processor group that is based on the processors included in proc_ids. This routine must be called by all processors in MPI_COMM_WORLD.

Parameters

interface	CommInterface object giving access to the MPI communication layer
proc_ids	set of ids that are to be integrated in the group. The ids number are to be understood in terms of MPI_COMM_WORLD ranks.

◆ MPIProcessorGroup() [3/6]

MEDCoupling::MPIProcessorGroup::MPIProcessorGroup	(	const CommInterface &	interface,
		std::map< std::string, std::set< int >>	proc_ids_by_name,
		const std::string &	simCodeTag,
		const MPI_Comm &	world_comm = `MPI_COMM_WORLD`
	)

◆ MPIProcessorGroup() [4/6]

MEDCoupling::MPIProcessorGroup::MPIProcessorGroup	(	const ProcessorGroup &	proc_group,
		std::set< int >	proc_ids
	)

◆ MPIProcessorGroup() [5/6]

MEDCoupling::MPIProcessorGroup::MPIProcessorGroup	(	const CommInterface &	comm_interface,
		int	pstart,
		int	pend,
		const MPI_Comm &	world_comm = `MPI_COMM_WORLD`
	)

Creates a processor group that is based on the processors between pstart and pend. This routine must be called by all processors in MPI_COMM_WORLD.

Parameters

comm_interface	CommInterface object giving access to the MPI communication layer
pstart	id in MPI_COMM_WORLD of the first processor in the group
pend	id in MPI_COMM_WORLD of the last processor in the group

References MEDCoupling::ProcessorGroup::_comm_interface, MEDCoupling::CommInterface::commCreate(), MEDCoupling::CommInterface::commGroup(), MEDCoupling::CommInterface::commRank(), MEDCoupling::CommInterface::commSize(), MEDCoupling::CommInterface::groupFree(), and MEDCoupling::CommInterface::groupIncl().

◆ MPIProcessorGroup() [6/6]

MEDCoupling::MPIProcessorGroup::MPIProcessorGroup ( const MPIProcessorGroup & other )

◆ ~MPIProcessorGroup()

MEDCoupling::MPIProcessorGroup::~MPIProcessorGroup ( )

virtual

References release().

Member Function Documentation

◆ release()

void MEDCoupling::MPIProcessorGroup::release ( )

Destructor involves MPI operations: make sure this is accessible from a proper method for Python wrapping.

References MEDCoupling::ProcessorGroup::_comm_interface, MEDCoupling::CommInterface::commFree(), and MEDCoupling::CommInterface::groupFree().

Referenced by ~MPIProcessorGroup().

◆ deepCopy()

MPIProcessorGroup * MEDCoupling::MPIProcessorGroup::deepCopy ( ) const

virtual

Implements MEDCoupling::ProcessorGroup.

References MPIProcessorGroup().

◆ fuse()

ProcessorGroup * MEDCoupling::MPIProcessorGroup::fuse ( const ProcessorGroup & group ) const

virtual

Adding processors of group group to local group.

Parameters

group group that is to be fused with current group

Returns: new group formed by the fusion of local group and group.

Implements MEDCoupling::ProcessorGroup.

References MEDCoupling::ProcessorGroup::_comm_interface, MEDCoupling::ProcessorGroup::_proc_ids, MEDCoupling::ProcessorGroup::getProcIDs(), and MPIProcessorGroup().

◆ intersect()

void MEDCoupling::MPIProcessorGroup::intersect ( ProcessorGroup & )

virtual

Implements MEDCoupling::ProcessorGroup.

◆ myRank()

int MEDCoupling::MPIProcessorGroup::myRank ( ) const

virtual

Implements MEDCoupling::ProcessorGroup.

Referenced by MEDCoupling::ExplicitCoincidentDEC::broadcastTopology().

◆ containsMyRank()

bool MEDCoupling::MPIProcessorGroup::containsMyRank ( ) const

virtual

Implements MEDCoupling::ProcessorGroup.

◆ translateRank()

int MEDCoupling::MPIProcessorGroup::translateRank	(	const ProcessorGroup *	group,
		int	rank
	)		const

virtual

Translation of the rank id between two processor groups. This method translates rank rank on the current processor group to the rank on group pointed by group.

Parameters