SST::Partition::SSTLinearPartition Class Reference

Performs a linear partition scheme of an SST simulation configuration. More...

#include <linpart.h>

Inheritance diagram for SST::Partition::SSTLinearPartition:

Public Member Functions
	SSTLinearPartition (RankInfo rankCount, int verbosity)
	Creates a new linear partition scheme. More...
void	performPartition (PartitionGraph *graph)
	Performs a partition of an SST simulation configuration. More...
bool	requiresConfigGraph ()
bool	spawnOnAllRanks ()
Public Member Functions inherited from SST::Partition::SSTPartitioner
virtual void	performPartition (ConfigGraph *graph)
	Function to be overriden by subclasses. More...

Static Public Member Functions
static SSTPartitioner *	allocate (RankInfo total_ranks, RankInfo my_rank, int verbosity)
Static Public Member Functions inherited from SST::Partition::SSTPartitioner
static bool	addPartitioner (const std::string name, const SSTPartitioner::partitionerAlloc alloc, const std::string description)
static SSTPartitioner *	getPartitioner (std::string name, RankInfo total_ranks, RankInfo my_rank, int verbosity)
static const std::map< std::string, std::string > &	getDescriptionMap ()

Protected Attributes
RankInfo	rankcount
	Number of ranks in the simulation.
Output *	partOutput
	Output object to print partitioning information.

Static Protected Attributes
static bool	initialized = SSTPartitioner::addPartitioner("linear",&SSTLinearPartition::allocate, "Partitions components by dividing Component ID space into roughly equal portions. Components with sequential IDs will be placed close together.")

Additional Inherited Members
Public Types inherited from SST::Partition::SSTPartitioner
typedef SSTPartitioner *(*	partitionerAlloc) (RankInfo total_ranks, RankInfo my_rank, int verbosity)

Detailed Description

Performs a linear partition scheme of an SST simulation configuration.

In this scheme a list of components (supplied as a graph) are grouped by slicing the list into approximately equal parts. A "part" is generated for each MPI rank performing the simulation. This means Components with sequential ids will be placed close together. In general this scheme provides a very effective partition for most parallel simulations which generate man similar components of interest close together in the input Python configuration. It is also very fast to compute a linear partiton scheme. For more aggressive partition schemes users should try either a simple or Zoltan-based partitioner.

Constructor & Destructor Documentation

SSTLinearPartition::SSTLinearPartition	(	RankInfo	rankCount,
		int	verbosity
	)

Creates a new linear partition scheme.

Parameters

mpiRankCount	Number of MPI ranks in the simulation
verbosity	The level of information to output

References SST::Output::STDOUT.

Member Function Documentation

void SSTLinearPartition::performPartition ( PartitionGraph *  graph )

virtual

Performs a partition of an SST simulation configuration.

Parameters

graph

The simulation configuration to partition

Reimplemented from SST::Partition::SSTPartitioner.

---

The documentation for this class was generated from the following files:

• src/sst/core/part/linpart.h
• src/sst/core/part/linpart.cc