Implements a MemoryManager that binds the thread responsible for the MemoryManager to a CUDA GPU prior to allocating memory. More...

#include <htgs/core/memory/CudaMemoryManager.hpp>

Inheritance diagram for htgs::CudaMemoryManager< T >:

Collaboration diagram for htgs::CudaMemoryManager< T >:

Public Member Functions
	CudaMemoryManager (std::string name, int *cudaIds, size_t memoryPoolSize, std::shared_ptr< IMemoryAllocator< T >> memoryAllocator, MMType type)
	Creates a CudaMemoryManager. More...

void	initialize () override
	Initializes the CudaMemoryManager by setting which GPU the CudaMemoryManager is repsonsible prior to allocating memory. More...

std::string	getName () override
	Gets the name of the CudaMemoryManager. More...

MemoryManager< T > *	copy () override
	Creates a shallow copy of the CudaMemoryManager. More...

Public Member Functions inherited from htgs::MemoryManager< T >
	MemoryManager (std::string name, size_t memoryPoolSize, std::shared_ptr< IMemoryAllocator< T >> memoryAllocator, MMType type)
	Creates the MemoryManager with the specified memory pool size and allocator. More...

	~MemoryManager () override
	Destructor.

void	initialize () override
	Initializes the MemoryManager, getting the size of the memory pool, and filling the memory pool with allocated data. More...

void	shutdown () override
	Shuts down the MemoryManager memory is only released when the underlying graph destructs the memory manager.

void	executeTask (std::shared_ptr< MemoryData< T >> data) override
	Processes memory data. More...

void	debug () override
	Provides debug output for MemoryManager.

virtual size_t	getMemoryPoolSize ()
	Virtual function to gets the size of the MemoryPool. More...

virtual std::shared_ptr< IMemoryAllocator< T > >	getAllocator ()
	Gets the allocator that is responsible for allocating and freeing memory for the MemoryPool. More...

std::string	getMemoryManagerName ()
	Gets the name of the memory manager. More...

std::string	typeName ()
	Gets the demangled type name of the connector. More...

MMType	getType () const
	Gets the memory manager type. More...

virtual std::string	genDot (int flags, std::string dotId, std::shared_ptr< AnyConnector > input, std::shared_ptr< AnyConnector > output) override
	Virtual function that generates the input/output and per-task dot notation. More...

std::string	getDotFillColor () override
	Gets the color for filling the shape for graphviz dot. More...

Public Member Functions inherited from htgs::ITask< MemoryData< T >, MemoryData< T > >
	ITask ()
	Creates an ITask with number of threads equal to 1.

	ITask (size_t numThreads)
	Constructs an ITask with a specified number of threads. More...

	ITask (size_t numThreads, bool isStartTask, bool poll, size_t microTimeoutTime)
	Constructs an ITask with a specified number of threads as well as additional scheduling options. More...

void	initialize (size_t pipelineId, size_t numPipeline, TaskManager< MemoryData< T >, MemoryData< T > > *ownerTask)
	Function that is called when an ITask is being initialized by it's owner thread. More...

virtual void	executeTask (std::shared_ptr< MemoryData< T > > data)=0
	Pure virtual function that is called when an ITask's thread is to execute on data. More...

virtual bool	canTerminate (std::shared_ptr< AnyConnector > inputConnector) override

virtual void	executeTaskFinal () override

virtual std::string	getDotLabelName () override

virtual std::string	getDotShapeColor () override

virtual std::string	getDotShape () override

virtual std::string	getDotCustomProfile () override
	Adds the string text to the profiling of this task in the graphviz dot visualization. More...

virtual void	printProfile () override

virtual size_t	getNumGraphsSpawned ()
	Gets the number of graphs spawned by this ITask. More...

virtual std::string	genDotProducerEdgeToTask (std::map< std::shared_ptr< AnyConnector >, AnyITask * > &inputConnectorDotMap, int dotFlags) override

virtual std::string	genDotConsumerEdgeFromConnector (std::shared_ptr< AnyConnector > connector, int flags) override

virtual std::string	genDotProducerEdgeFromConnector (std::shared_ptr< AnyConnector > connector, int flags)

ITask< MemoryData< T >, MemoryData< T > > *	copyITask (bool deep) override
	Copies the ITask (including a copy of all memory edges) More...

void	addResult (std::shared_ptr< MemoryData< T > > result)
	Adds results to the output list to be sent to the next connected ITask in a TaskGraph. More...

void	addResult (MemoryData< T > *result)
	Adds results to the output list to be sent to the next connected ITask in a TaskGraph. More...

m_data_t< V >	getMemory (std::string name, IMemoryReleaseRule *releaseRule)
	Retrieves memory from a memory edge. More...

m_data_t< V >	getDynamicMemory (std::string name, IMemoryReleaseRule *releaseRule, size_t numElems)
	Retrieves memory from a memory edge. More...

void	releaseMemory (m_data_t< V > memory)
	Releases memory onto a memory edge, which is transferred by the graph communicator. More...

void	resetProfile ()
	Resets profile data.

size_t	getThreadID ()
	Gets the thread ID associated with this task. More...

unsigned long long int	getTaskComputeTime () const
	Gets the task's compute time. More...

std::string	inTypeName () override final

std::string	outTypeName () override final

std::string	getAddress () override final

void	setTaskManager (TaskManager< MemoryData< T >, MemoryData< T > > *ownerTask)
	Sets the owner task manager for this ITask. More...

TaskManager< MemoryData< T >, MemoryData< T > > *	getOwnerTaskManager ()
	Gets the owner task manager for this ITask. More...

virtual void	gatherProfileData (std::map< AnyTaskManager , TaskManagerProfile > *taskManagerProfiles)
	Gathers profile data. More...

Public Member Functions inherited from htgs::AnyITask
	AnyITask ()
	Creates an ITask with number of threads equal to 1.

	AnyITask (size_t numThreads)
	Constructs an ITask with a specified number of threads. More...

	AnyITask (size_t numThreads, bool isStartTask, bool poll, size_t microTimeoutTime)
	Constructs an ITask with a specified number of threads as well as additional scheduling options. More...

virtual	~AnyITask ()
	Destructor.

virtual std::string	genDot (int flags, std::string dotId, std::shared_ptr< htgs::AnyConnector > input, std::shared_ptr< htgs::AnyConnector > output)
	Virtual function that generates the input/output and per-task dot notation. More...

virtual std::string	getConsumerDotIds ()

virtual std::string	getProducerDotIds ()

virtual std::string	genDot (int flags, std::string dotId)
	Virtual function that adds additional dot attributes to this node. More...

virtual std::string	genCustomDot (ProfileUtils *profileUtils, int colorFlag)
	Virtual function to generate customized dot file. More...

virtual std::string	debugDotNode ()
	Provides debug output for a node in the dot graph. More...

virtual void	profile ()
	Virtual function that is called to provide profile output for the ITask. More...

virtual std::string	profileStr ()
	Virtual function that is called after executionTask is called. More...

void	initialize (size_t pipelineId, size_t numPipeline)
	Virtual function that is called when an ITask is being initialized by it's owner thread. More...

void	setPipelineId (size_t pipelineId)
	Sets the pipeline Id for this ITask. More...

size_t	getPipelineId ()
	Gets the pipeline ID. More...

void	setNumPipelines (size_t numPipelines)
	Sets the number of pipelines that this ITask belongs too. More...

size_t	getNumPipelines () const
	Sets the task graph communicator. More...

size_t	getNumThreads () const
	Gets the number of threads associated with this ITask. More...

bool	isStartTask () const
	Gets whether this ITask is a starting task. More...

bool	isPoll () const
	Gets whether this ITask is polling for data or not. More...

size_t	getMicroTimeoutTime () const
	Gets the timeout time for polling. More...

void	copyMemoryEdges (AnyITask *iTaskCopy)
	Copies the memory edges from this AnyITask to another AnyITask. More...

std::string	genDot (int flags, std::shared_ptr< AnyConnector > input, std::shared_ptr< AnyConnector > output)
	Creates a dot notation representation for this task. More...

void	profileITask ()
	Provides profile output for the ITask,. More...

std::string	getDotId ()
	Gets the id used for dot nodes. More...

std::string	getNameWithPipelineId ()
	Gets the name of the ITask with it's pipeline ID. More...

const std::shared_ptr< ConnectorMap > &	getMemoryEdges () const
	Gets the memory edges for the task. More...

const std::shared_ptr< ConnectorMap > &	getReleaseMemoryEdges () const
	Gets the memory edges for releasing memory for the memory manager, used to shutdown the memory manager. More...

bool	hasMemoryEdge (std::string name)
	Checks whether this ITask contains a memory edge for a specified name. More...

void	attachMemoryEdge (std::string name, std::shared_ptr< AnyConnector > getMemoryConnector, std::shared_ptr< AnyConnector > releaseMemoryConnector, MMType type)
	Attaches a memory edge to this ITask to get memory. More...

unsigned long long int	getMemoryWaitTime () const
	Gets the amount of time the task was waiting for memory. More...

void	incMemoryWaitTime (unsigned long long int val)
	Increments memory wait time. More...

Private Attributes
int *	cudaIds
	The array of CUDA contexts.

Detailed Description

template<class T>
class htgs::CudaMemoryManager< T >

Implements a MemoryManager that binds the thread responsible for the MemoryManager to a CUDA GPU prior to allocating memory.

Once a TaskGraphRuntime binds a thread to the CudaMemoryManager and calls its initialize function, the CUDA GPU specified by the pipelineId of the CudaMemoryManager is bound to the thread. This pipelineId accesses a Cuda ID, so the number of pipelines spawned for the ExecutionPipeline task should match the number of Cuda IDs passed to the CudaMemoryManager If the Task is not associated with an ExecutionPipeline, then there only needs to be one Cuda ID.

Template Parameters

T	the input/output MemoryData type for the CudaMemoryManager; i.e.; cufftDoubleComplex

Constructor & Destructor Documentation

◆ CudaMemoryManager()

template<class T>

htgs::CudaMemoryManager< T >::CudaMemoryManager	(	std::string	name,
		int *	cudaIds,
		size_t	memoryPoolSize,
		std::shared_ptr< IMemoryAllocator< T >>	memoryAllocator,
		MMType	type
	)

inline

Creates a CudaMemoryManager.

The CUcontext should contain enough CUcontext such that there is one per Cuda GPU if this Task is added into an ExecutionPipeline.

Parameters

name	the name of the memory manager edge
cudaIds	the Cuda Ids
memoryPoolSize	the size of the memory pool
memoryAllocator	the memory allocator describing how memory is allocated for the GPU.
type	the memory manager type

Member Function Documentation

◆ copy()

template<class T>

MemoryManager<T>* htgs::CudaMemoryManager< T >::copy ( )

inlineoverridevirtual

Creates a shallow copy of the CudaMemoryManager.

Returns: the copy of the CudaMemoryManager

Reimplemented from htgs::MemoryManager< T >.

◆ getName()

template<class T>

std::string htgs::CudaMemoryManager< T >::getName ( )

inlineoverridevirtual

Gets the name of the CudaMemoryManager.

Returns

Reimplemented from htgs::MemoryManager< T >.

◆ initialize()

template<class T>

void htgs::CudaMemoryManager< T >::initialize ( )

inlineoverridevirtual

Initializes the CudaMemoryManager by setting which GPU the CudaMemoryManager is repsonsible prior to allocating memory.

The initialize routine is called after a thread has been bound to the Task, thus enforcing the Task to allocate memory on the specified Cuda GPU based on the pipelineId associated with the Task managing the CudaMemoryManager.

Reimplemented from htgs::ITask< MemoryData< T >, MemoryData< T > >.

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

/home/tjb3/work/HTGS/src/htgs/core/memory/CudaMemoryManager.hpp

Public Member Functions

Private Attributes

Detailed Description

template<class T> class htgs::CudaMemoryManager< T >

Constructor & Destructor Documentation

◆ CudaMemoryManager()

Member Function Documentation

◆ copy()

◆ getName()

◆ initialize()

template<class T>
class htgs::CudaMemoryManager< T >