autopas::MPIParallelizedStrategy Class Reference

This strategy spreads the configuration queue in a round robin fashion over all ranks (with similar domain). More...

#include <MPIParallelizedStrategy.h>

Inheritance diagram for autopas::MPIParallelizedStrategy:

Collaboration diagram for autopas::MPIParallelizedStrategy:

Public Member Functions
	MPIParallelizedStrategy (const Configuration &fallbackConfiguration, const AutoPas_MPI_Comm &comm, double mpiTuningMaxDifferenceForBucket, double mpiTuningWeightForMaxDensity)
	Constructor.
TuningStrategyOption	getOptionType () const override
	Get this object's associated TuningStrategyOption type.
bool	optimizeSuggestions (std::vector< Configuration > &configQueue, const EvidenceCollection &evidenceCollection) override
	Optimizes the queue of configurations to process.
void	receiveDomainSimilarityStatistics (double pdBinStdDevDensity, double pdBinMaxDensity) override
	Method to pass smoothed domain similarity statistics (particle-dependent bin standard deviation in density and max density) to the tuning strategy.
bool	reset (size_t iteration, size_t tuningPhase, std::vector< Configuration > &configQueue, const autopas::EvidenceCollection &evidenceCollection) override
	Reset all internal parameters to the beginning of a new tuning phase.
bool	needsDomainSimilarityStatistics () const override
	Indicate whether the strategy needs domain similarity statistics.
void	rejectConfiguration (const Configuration &configuration, bool indefinitely) override
	Notify the strategy about a configuration that is (currently) invalid and thus can potentially be dropped from some internal storage.
const AutoPas_MPI_Comm &	getBucket () const
	Get the MPI communicator for this rank's bucket.
Public Member Functions inherited from autopas::TuningStrategyInterface
virtual TuningStrategyOption	getOptionType () const =0
	Get this object's associated TuningStrategyOption type.
virtual void	addEvidence (const Configuration &configuration, const Evidence &evidence)
	Notifies the strategy about empirically collected information for the given configuration.
virtual bool	optimizeSuggestions (std::vector< Configuration > &configQueue, const EvidenceCollection &evidenceCollection)=0
	Optimizes the queue of configurations to process.
virtual bool	reset (size_t iteration, size_t tuningPhase, std::vector< Configuration > &configQueue, const autopas::EvidenceCollection &evidenceCollection)=0
	Reset all internal parameters to the beginning of a new tuning phase.
virtual bool	needsLiveInfo () const
	Returns whether this tuning strategy wants to get a LiveInfo object passed before a new tuning phase.
virtual void	receiveLiveInfo (const LiveInfo &info)
	Virtual method that subclasses can override to receive the LiveInfo object before a tuning phase if they return true in needsLiveInfo().
virtual void	rejectConfiguration (const Configuration &configuration, bool indefinitely)
	Notify the strategy about a configuration that is (currently) invalid and thus can potentially be dropped from some internal storage.
virtual bool	needsDomainSimilarityStatistics () const
	Indicate whether the strategy needs domain similarity statistics.
virtual void	receiveDomainSimilarityStatistics (double pdBinStdDevDensity, double pdBinMaxDensity)
	Method to pass smoothed domain similarity statistics (particle-dependent bin standard deviation in density and max density) to the tuning strategy.

Static Public Member Functions
static Configuration	createFallBackConfiguration (const std::set< Configuration > &searchSpace, const InteractionTypeOption &interactionType)
	Create a resilient configuration that should always be applicable.

Detailed Description

This strategy spreads the configuration queue in a round robin fashion over all ranks (with similar domain).

The actual splitting of the search space and details of the communication logic is not currently handled by this class, but by AutoPasConfigurationCommunicator.

Constructor & Destructor Documentation

MPIParallelizedStrategy()

autopas::MPIParallelizedStrategy::MPIParallelizedStrategy	(	const Configuration &	fallbackConfiguration,
		const AutoPas_MPI_Comm &	comm,
		double	mpiTuningMaxDifferenceForBucket,
		double	mpiTuningWeightForMaxDensity
	)

Constructor.

Parameters

fallbackConfiguration	Generally applicable configuration that can be used if no local configuration are applicable. It is suggested to use MPIParallelizedStrategy::createFallBackConfiguration() for this.
comm	The communicator holding all ranks which participate in this tuning strategy.
mpiTuningMaxDifferenceForBucket
mpiTuningWeightForMaxDensity

Member Function Documentation

createFallBackConfiguration()

Configuration autopas::MPIParallelizedStrategy::createFallBackConfiguration ( const std::set< Configuration > &  searchSpace, const InteractionTypeOption &  interactionType  )

static

Create a resilient configuration that should always be applicable.

The idea is to pick some configuration that is basically always applicable (lc_c08) and guess from the search space which options Newton3 and Data Layout options are allowed.

Parameters

searchSpace
interactionType

Returns

The fall back configuration.

getBucket()

const AutoPas_MPI_Comm & autopas::MPIParallelizedStrategy::getBucket

(

)

const

Get the MPI communicator for this rank's bucket.

Returns

getOptionType()

TuningStrategyOption autopas::MPIParallelizedStrategy::getOptionType ( ) const

overridevirtual

Get this object's associated TuningStrategyOption type.

Returns

TuningStrategyOption

Implements autopas::TuningStrategyInterface.

needsDomainSimilarityStatistics()

bool autopas::MPIParallelizedStrategy::needsDomainSimilarityStatistics ( ) const

inlineoverridevirtual

Indicate whether the strategy needs domain similarity statistics.

Returns

Reimplemented from autopas::TuningStrategyInterface.

optimizeSuggestions()

bool autopas::MPIParallelizedStrategy::optimizeSuggestions ( std::vector< Configuration > &  configQueue, const EvidenceCollection &  evidenceCollection  )

overridevirtual

Optimizes the queue of configurations to process.

This function is called once before each iteration in a tuning phase so all tuning strategies can give their input on which configuration to try next. This is done by reordering configQueue so that the next configuration to try is at the end (FIFO).

Parameters

configQueue	Queue of configurations to be tested. The tuning strategy should edit this queue.
evidenceCollection	All collected evidence until now.

Returns

boolean value to signal if the tuning strategy has intentionally wiped the config queue

Implements autopas::TuningStrategyInterface.

receiveDomainSimilarityStatistics()

void autopas::MPIParallelizedStrategy::receiveDomainSimilarityStatistics ( double  pdBinStdDevDensity, double  pdBinMaxDensity  )

overridevirtual

Method to pass smoothed domain similarity statistics (particle-dependent bin standard deviation in density and max density) to the tuning strategy.

See LiveInfo::gather for meaning of particle dependent bin.

Parameters

pdBinStdDevDensity	particle-dependent bin density standard deviation.
pdBinMaxDensity	particle-dependent bin maximum density.

Reimplemented from autopas::TuningStrategyInterface.

rejectConfiguration()

void autopas::MPIParallelizedStrategy::rejectConfiguration ( const Configuration &  configuration, bool  indefinitely  )

overridevirtual

Notify the strategy about a configuration that is (currently) invalid and thus can potentially be dropped from some internal storage.

Parameters

configuration
indefinitely	Whether the given configuration will never be valid

Reimplemented from autopas::TuningStrategyInterface.

reset()

bool autopas::MPIParallelizedStrategy::reset ( size_t  iteration, size_t  tuningPhase, std::vector< Configuration > &  configQueue, const autopas::EvidenceCollection &  evidenceCollection  )

overridevirtual

Reset all internal parameters to the beginning of a new tuning phase.

This can also mean to reorder the configQueue to some initially expected state.

Parameters

iteration	Gives the current iteration to the tuning strategy.
tuningPhase	Gives the current tuning phase to the tuning strategy.
configQueue	Queue of configurations to be tested. The tuning strategy should edit this queue.
evidenceCollection	All collected evidence until now.

Returns

boolean value to signal if the tuning strategy has intentionally wiped the config queue

Implements autopas::TuningStrategyInterface.

---

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

• autopas/tuning/tuningStrategy/MPIParallelizedStrategy.h
• autopas/tuning/tuningStrategy/MPIParallelizedStrategy.cpp