Processor Groups

Functions
	ParaMEDMEM::MPIProcessorGroup.MPIProcessorGroup (const CommInterface &interface)
	ParaMEDMEM::MPIProcessorGroup.MPIProcessorGroup (const CommInterface &interface, std::set< int > proc_ids, const MPI_Comm &world_comm=MPI_COMM_WORLD)
	ParaMEDMEM::MPIProcessorGroup.MPIProcessorGroup (const CommInterface &interface, int pstart, int pend, const MPI_Comm &world_comm=MPI_COMM_WORLD)
int	ParaMEDMEM::MPIProcessorGroup.translateRank (const ProcessorGroup *group, int rank) const
ProcessorGroup *	ParaMEDMEM::MPIProcessorGroup.createComplementProcGroup () const
virtual ProcessorGroup *	ParaMEDMEM::MPIProcessorGroup.fuse (const ProcessorGroup &) const
int	ParaMEDMEM::MPIProcessorGroup.myRank () const

---

Detailed Description

Overview

The MPIProcessorGroup class is used to set up processor groups that help to define the MPI topology of the couplings. They 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);
 MPIProcessorGroup codeB_group(comm_interface, 3, 4);
   
 ...
 }
 

---

Function Documentation

ParaMEDMEM.MPIProcessorGroup::MPIProcessorGroup

(

const CommInterface &

interface

)

[inherited]

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

Parameters:

interface

CommInterface object giving access to the MPI communication layer

References ParaMEDMEM.MPIProcessorGroup._comm, ParaMEDMEM.ProcessorGroup._comm_interface, ParaMEDMEM.MPIProcessorGroup._group, ParaMEDMEM.ProcessorGroup._proc_ids, ParaMEDMEM.MPIProcessorGroup._world_comm, ParaMEDMEM.CommInterface.commGroup(), ParaMEDMEM.CommInterface.commSize(), and ParaMEDMEM.ProcessorGroup.size().

ParaMEDMEM.MPIProcessorGroup::MPIProcessorGroup	(	const CommInterface &	interface,
		std::set< int >	proc_ids,
		const MPI_Comm &	world_comm = MPI_COMM_WORLD
	)		[inherited]

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.

References ParaMEDMEM.MPIProcessorGroup._comm, ParaMEDMEM.ProcessorGroup._comm_interface, ParaMEDMEM.MPIProcessorGroup._group, ParaMEDMEM.MPIProcessorGroup._world_comm, ParaMEDMEM.CommInterface.commCreate(), ParaMEDMEM.CommInterface.commGroup(), ParaMEDMEM.CommInterface.commRank(), ParaMEDMEM.CommInterface.commSize(), and ParaMEDMEM.CommInterface.groupIncl().

ParaMEDMEM.MPIProcessorGroup::MPIProcessorGroup	(	const CommInterface &	comm_interface,
		int	pstart,
		int	pend,
		const MPI_Comm &	world_comm = MPI_COMM_WORLD
	)		[inherited]

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 ParaMEDMEM.MPIProcessorGroup._comm, ParaMEDMEM.ProcessorGroup._comm_interface, ParaMEDMEM.MPIProcessorGroup._group, ParaMEDMEM.MPIProcessorGroup._world_comm, ParaMEDMEM.CommInterface.commCreate(), ParaMEDMEM.CommInterface.commGroup(), ParaMEDMEM.CommInterface.commRank(), ParaMEDMEM.CommInterface.commSize(), and ParaMEDMEM.CommInterface.groupIncl().

int ParaMEDMEM.MPIProcessorGroup::translateRank	(	const ProcessorGroup *	group,
		int	rank
	)		const [virtual, inherited]

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:

group	group from which the rank is expected
rank	rank on group group of the processor which is to be translated

Returns:

rank on local group

Implements ParaMEDMEM.ProcessorGroup.

References ParaMEDMEM.MPIProcessorGroup._group, and med_test1.group.

ProcessorGroup * ParaMEDMEM.MPIProcessorGroup::createComplementProcGroup

(

)

const [virtual, inherited]

Creates a processor group that is the complement of the current group inside MPI_COMM_WORLD

Returns:

pointer to the new ProcessorGroup structure.

Implements ParaMEDMEM.ProcessorGroup.

References ParaMEDMEM.ProcessorGroup._comm_interface, ParaMEDMEM.ProcessorGroup._proc_ids, ParaMEDMEM.MPIProcessorGroup._world_comm, ParaMEDMEM.MPIProcessorGroup.MPIProcessorGroup(), and ParaMEDMEM.CommInterface.worldSize().

ProcessorGroup * ParaMEDMEM.MPIProcessorGroup::fuse

(

const ProcessorGroup &

group

)

const [virtual, inherited]

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 ParaMEDMEM.ProcessorGroup.

References ParaMEDMEM.ProcessorGroup._comm_interface, ParaMEDMEM.ProcessorGroup._proc_ids, ParaMEDMEM.MPIProcessorGroup._world_comm, ParaMEDMEM.ProcessorGroup.getProcIDs(), and ParaMEDMEM.MPIProcessorGroup.MPIProcessorGroup().

int ParaMEDMEM.MPIProcessorGroup::myRank

(

)

const [virtual, inherited]

Implements ParaMEDMEM.ProcessorGroup.

References ParaMEDMEM.MPIProcessorGroup._comm, and test_NonCoincidentDEC.rank.