The compute service base class. More...

#include <ComputeService.h>

Inheritance diagram for wrench::ComputeService:

Public Types
enum	TerminationCause { TERMINATION_NONE, TERMINATION_COMPUTE_SERVICE_TERMINATED, TERMINATION_JOB_KILLED, TERMINATION_JOB_TIMEOUT }
	Job termination cause enum.

Public Types inherited from wrench::S4U_Daemon
enum	State { UP, DOWN, SUSPENDED }
	Daemon states. More...

Public Member Functions
	ComputeService (const std::string &hostname, std::string service_name, std::string scratch_space_mount_point)
	Constructor. More...

std::map< std::string, double >	getCoreFlopRate ()
	Get the per-core flop rate of the compute service's hosts. More...

double	getFreeScratchSpaceSize ()
	Get the free space on the compute service's scratch storage space. More...

std::vector< std::string >	getHosts ()
	Get the list of the compute service's compute host. More...

std::map< std::string, double >	getMemoryCapacity ()
	Get the RAM capacities for each of the compute service's hosts. More...

unsigned long	getNumHosts ()
	Get the number of hosts that the compute service manages. More...

std::map< std::string, double >	getPerHostAvailableMemoryCapacity ()
	Get ram availability for each of the compute service's host. More...

std::map< std::string, unsigned long >	getPerHostNumCores ()
	Get core counts for each of the compute service's host. More...

std::map< std::string, unsigned long >	getPerHostNumIdleCores ()
	Get idle core counts for each of the compute service's host. More...

std::shared_ptr< StorageService >	getScratch ()
	Get the compute service's scratch storage space. More...

unsigned long	getTotalNumCores ()
	Get the total core counts for all hosts of the compute service. More...

virtual unsigned long	getTotalNumIdleCores ()
	Get the total idle core count for all hosts of the compute service. Note that this doesn't mean that asking for these cores right will mean immediate execution (since jobs may be pending and "ahead" in the queue, e.g., because they depend on current actions that are not using all available resources). More...

double	getTotalScratchSpaceSize ()
	Get the total capacity of the compute service's scratch storage space. More...

double	getTTL ()
	Get the time-to-live of the compute service. More...

virtual bool	hasScratch () const
	Checks if the compute service has a scratch space. More...

virtual bool	isThereAtLeastOneHostWithIdleResources (unsigned long num_cores, double ram)
	Method to find out if, right now, the compute service has at least one host with some idle number of cores and some available RAM. Note that this doesn't mean that asking for these resources right will mean immediate execution (since jobs may be pending and "ahead" in the queue, e.g., because they depend on current actions that are not using all available resources). More...

void	stop () override
	Stop the compute service.

virtual void	stop (bool send_failure_notifications, ComputeService::TerminationCause termination_cause)
	Stop the compute service. More...

virtual void	submitCompoundJob (std::shared_ptr< CompoundJob > job, const std::map< std::string, std::string > &service_specific_arguments)=0
	Method to submit a compound job to the service. More...

virtual bool	supportsCompoundJobs ()=0
	Returns true if the service supports pilot jobs. More...

virtual bool	supportsPilotJobs ()=0
	Returns true if the service supports compound jobs. More...

virtual bool	supportsStandardJobs ()=0
	Returns true if the service supports standard jobs. More...

virtual void	terminateCompoundJob (std::shared_ptr< CompoundJob > job)=0
	Method to terminate a compound job. More...

void	terminateJob (std::shared_ptr< CompoundJob > job)
	Terminate a previously-submitted job (which may or may not be running yet) More...

Public Member Functions inherited from wrench::Service
void	assertServiceIsUp ()
	Throws an exception if the service is not up. More...

std::string	getHostname ()
	Get the name of the host on which the service is / will be running. More...

const WRENCH_MESSAGE_PAYLOADCOLLECTION_TYPE &	getMessagePayloadList () const
	Get all message payloads and their values of the Service. More...

double	getMessagePayloadValue (WRENCH_MESSAGEPAYLOAD_TYPE)
	Get a message payload of the Service as a double. More...

double	getNetworkTimeoutValue ()
	Returns the service's network timeout value. More...

std::string	getPhysicalHostname ()
	Get the physical name of the host on which the service is / will be running. More...

bool	getPropertyValueAsBoolean (WRENCH_PROPERTY_TYPE)
	Get a property of the Service as a boolean. More...

double	getPropertyValueAsDouble (WRENCH_PROPERTY_TYPE)
	Get a property of the Service as a double. More...

std::string	getPropertyValueAsString (WRENCH_PROPERTY_TYPE)
	Get a property of the Service as a string. More...

unsigned long	getPropertyValueAsUnsignedLong (WRENCH_PROPERTY_TYPE)
	Get a property of the Service as an unsigned long. More...

bool	isUp ()
	Returns true if the service is UP, false otherwise. More...

void	resume ()
	Resume the service. More...

void	setNetworkTimeoutValue (double value)
	Sets the service's network timeout value. More...

void	setStateToDown ()
	Set the state of the service to DOWN.

void	start (std::shared_ptr< Service > this_service, bool daemonize, bool auto_restart)
	Start the service. More...

void	suspend ()
	Suspend the service.

Public Member Functions inherited from wrench::S4U_Daemon
	S4U_Daemon (std::string hostname, std::string process_name_prefix)
	Constructor (daemon with a mailbox) More...

void	acquireDaemonLock ()
	Method to acquire the daemon's lock. More...

virtual void	cleanup (bool has_returned_from_main, int return_value)
	Cleanup function called when the daemon terminates (for whatever reason). The default behavior is to throw an exception if the host is off. This method should be overriden in a daemons implements some fault-tolerant behavior, or is naturally tolerant. More...

void	createLifeSaver (std::shared_ptr< S4U_Daemon > reference)
	Create a life saver for the daemon. More...

std::string	getName ()
	Retrieve the process name. More...

int	getReturnValue ()
	Returns the value returned by main() (if the daemon has returned from main) More...

Simulation *	getSimulation ()
	Get the service's simulation. More...

S4U_Daemon::State	getState ()
	Get the daemon's state. More...

bool	hasReturnedFromMain ()
	Returns true if the daemon has returned from main() (i.e., not brutally killed) More...

bool	isDaemonized ()
	Return the daemonized status of the daemon. More...

bool	isSetToAutoRestart ()
	Return the auto-restart status of the daemon. More...

std::pair< bool, int >	join ()
	Join (i.e., wait for) the daemon. More...

virtual int	main ()=0
	The daemon's main method, to be overridden. More...

void	releaseDaemonLock ()
	Method to release the daemon's lock. More...

void	resumeActor ()
	Resume the daemon/actor.

void	setSimulation (Simulation *simulation)
	Set the service's simulation. More...

void	setupOnExitFunction ()
	Sets up the on_exit function for the actor.

void	startDaemon (bool _daemonized, bool _auto_restart)
	Start the daemon. More...

void	suspendActor ()
	Suspend the daemon/actor.

Static Public Attributes
static constexpr unsigned long	ALL_CORES = ULONG_MAX
	A convenient constant to mean "use all cores of a physical host" whenever a number of cores is needed when instantiating compute services.

static constexpr double	ALL_RAM = DBL_MAX
	A convenient constant to mean "use all ram of a physical host" whenever a ram capacity is needed when instantiating compute services.

Static Public Attributes inherited from wrench::S4U_Daemon
static std::unordered_map< aid_t, simgrid::s4u::Mailbox * >	map_actor_to_recv_mailbox

Protected Member Functions
	ComputeService (const std::string &hostname, std::string service_name, std::shared_ptr< StorageService > scratch_space)
	Constructor. More...

std::shared_ptr< StorageService >	getScratchSharedPtr ()
	Get a shared pointer to the compute service's scratch storage space. More...

void	submitJob (std::shared_ptr< CompoundJob > job, const std::map< std::string, std::string > &={})
	Submit a job to the compute service. More...

virtual void	validateJobsUseOfScratch (std::map< std::string, std::string > &service_specific_args)
	Method to validate that a job's use of the scratch space if ok. Throws exception if not. More...

virtual void	validateServiceSpecificArguments (std::shared_ptr< CompoundJob > job, std::map< std::string, std::string > &service_specific_args)
	Method the validates service-specific arguments (throws std::invalid_argument if invalid) More...

Protected Member Functions inherited from wrench::Service
	Service (std::string hostname, std::string process_name_prefix)
	Constructor. More...

	~Service () override
	Destructor.

template<class T >
std::shared_ptr< T >	getSharedPtr ()
	Method to retrieve the shared_ptr to a service. More...

void	serviceSanityCheck ()
	Check whether the service is properly configured and running. More...

void	setMessagePayload (WRENCH_MESSAGEPAYLOAD_TYPE, double)
	Set a message payload of the Service. More...

void	setMessagePayloads (WRENCH_MESSAGE_PAYLOADCOLLECTION_TYPE default_messagepayload_values, WRENCH_MESSAGE_PAYLOADCOLLECTION_TYPE overriden_messagepayload_values)
	Set default and user-defined message payloads. More...

void	setProperties (WRENCH_PROPERTY_COLLECTION_TYPE default_property_values, WRENCH_PROPERTY_COLLECTION_TYPE overriden_property_values)
	Set default and user-defined properties. More...

void	setProperty (WRENCH_PROPERTY_TYPE, const std::string &)
	Set a property of the Service. More...

Protected Member Functions inherited from wrench::S4U_Daemon
bool	killActor ()
	Kill the daemon/actor (does nothing if already dead) More...

void	runMainMethod ()
	Method that run's the user-defined main method (that's called by the S4U actor class)

Protected Attributes
std::shared_ptr< StorageService >	scratch_space_storage_service
	A scratch storage service associated to the compute service.

Protected Attributes inherited from wrench::Service
WRENCH_MESSAGE_PAYLOADCOLLECTION_TYPE	messagepayload_list
	The service's messagepayload list.

std::string	name
	The service's name.

double	network_timeout = 30.0
	The time (in seconds) after which a service that doesn't send back a reply (control) message causes a NetworkTimeOut exception. (default: 30 second; if <0 never timeout)

WRENCH_PROPERTY_COLLECTION_TYPE	property_list
	The service's property list.

bool	shutting_down = false
	A boolean that indicates if the service is in the middle of shutting down.

Protected Attributes inherited from wrench::S4U_Daemon
unsigned int	num_starts = 0
	The number of time that this daemon has started (i.e., 1 + number of restarts)

Simulation *	simulation
	a pointer to the simulation object

State	state
	The service's state.

Additional Inherited Members
Static Public Member Functions inherited from wrench::S4U_Daemon
static simgrid::s4u::Mailbox *	getRunningActorRecvMailbox ()
	Return the running actor's recv mailbox. More...

Public Attributes inherited from wrench::S4U_Daemon
std::string	hostname
	The name of the host on which the daemon is running.

LifeSaver *	life_saver = nullptr
	The daemon's life saver.

simgrid::s4u::Mailbox *	mailbox
	The daemon's mailbox.

std::string	process_name
	The name of the daemon.

simgrid::s4u::Mailbox *	recv_mailbox
	The daemon's receive mailbox (to send to another daemon so that that daemon can reply)

Static Protected Member Functions inherited from wrench::Service
static void	assertServiceIsUp (std::shared_ptr< Service > s)
	Assert for the service being up. More...

Detailed Description

The compute service base class.

Constructor & Destructor Documentation

◆ ComputeService() [1/2]

wrench::ComputeService::ComputeService	(	const std::string &	hostname,
		std::string	service_name,
		std::string	scratch_space_mount_point
	)

Constructor.

Parameters

hostname	the name of the host on which the compute service runs
service_name	the name of the compute service
mailbox_name_prefix	the mailbox name prefix
scratch_space_mount_point	the service's scratch space's mount point ("" if none)

◆ ComputeService() [2/2]

wrench::ComputeService::ComputeService	(	const std::string &	hostname,
		std::string	service_name,
		std::shared_ptr< StorageService >	scratch_space
	)

protected

Constructor.

Parameters

hostname	the name of the host on which the compute service runs
service_name	the name of the compute service
mailbox_name_prefix	the mailbox name prefix
scratch_space	scratch storage space of the compute service (nullptr if none)

Member Function Documentation

◆ getCoreFlopRate()

std::map< std::string, double > wrench::ComputeService::getCoreFlopRate ( )

Get the per-core flop rate of the compute service's hosts.

Returns: a list of flop rates in flop/sec

Exceptions

ExecutionException

◆ getFreeScratchSpaceSize()

double wrench::ComputeService::getFreeScratchSpaceSize ( )

Get the free space on the compute service's scratch storage space.

Returns: a size (in bytes)

◆ getHosts()

std::vector< std::string > wrench::ComputeService::getHosts ( )

Get the list of the compute service's compute host.

Returns: a vector of hostnames

Exceptions

ExecutionException
std::runtime_error

◆ getMemoryCapacity()

std::map< std::string, double > wrench::ComputeService::getMemoryCapacity ( )

Get the RAM capacities for each of the compute service's hosts.

Returns: a map of RAM capacities, indexed by hostname

Exceptions

ExecutionException

◆ getNumHosts()

unsigned long wrench::ComputeService::getNumHosts ( )

Get the number of hosts that the compute service manages.

Returns: the host count

Exceptions

ExecutionException
std::runtime_error

◆ getPerHostAvailableMemoryCapacity()

std::map< std::string, double > wrench::ComputeService::getPerHostAvailableMemoryCapacity ( )

Get ram availability for each of the compute service's host.

Returns: the ram availability map (could be empty)

Exceptions

ExecutionException
std::runtime_error

◆ getPerHostNumCores()

std::map< std::string, unsigned long > wrench::ComputeService::getPerHostNumCores ( )

Get core counts for each of the compute service's host.

Returns: a map of core counts, indexed by hostnames

Exceptions

ExecutionException
std::runtime_error

◆ getPerHostNumIdleCores()

std::map< std::string, unsigned long > wrench::ComputeService::getPerHostNumIdleCores ( )

Get idle core counts for each of the compute service's host.

Returns: the idle core counts (could be empty). Note that this doesn't mean that asking for these cores right will mean immediate execution (since jobs may be pending and "ahead" in the queue, e.g., because they depend on current actions that are not using all available resources).

Exceptions

ExecutionException
std::runtime_error

◆ getScratch()

std::shared_ptr< StorageService > wrench::ComputeService::getScratch ( )

Get the compute service's scratch storage space.

Returns: a pointer to the shared scratch space

◆ getScratchSharedPtr()

std::shared_ptr< StorageService > wrench::ComputeService::getScratchSharedPtr ( )

protected

Get a shared pointer to the compute service's scratch storage space.

Returns: a shared pointer to the shared scratch space

◆ getTotalNumCores()

unsigned long wrench::ComputeService::getTotalNumCores ( )

Get the total core counts for all hosts of the compute service.

Returns: total core counts

Exceptions

ExecutionException
std::runtime_error

◆ getTotalNumIdleCores()

unsigned long wrench::ComputeService::getTotalNumIdleCores ( )

virtual

Get the total idle core count for all hosts of the compute service. Note that this doesn't mean that asking for these cores right will mean immediate execution (since jobs may be pending and "ahead" in the queue, e.g., because they depend on current actions that are not using all available resources).

Returns: total idle core count.

Exceptions

ExecutionException
std::runtime_error

◆ getTotalScratchSpaceSize()

double wrench::ComputeService::getTotalScratchSpaceSize ( )

Get the total capacity of the compute service's scratch storage space.

Returns: a size (in bytes)

◆ getTTL()

double wrench::ComputeService::getTTL ( )

Get the time-to-live of the compute service.

Returns: the ttl in seconds

Exceptions

ExecutionException

◆ hasScratch()

bool wrench::ComputeService::hasScratch ( ) const

virtual

Checks if the compute service has a scratch space.

Returns: true if the compute service has some scratch storage space, false otherwise

◆ isThereAtLeastOneHostWithIdleResources()

bool wrench::ComputeService::isThereAtLeastOneHostWithIdleResources	(	unsigned long	num_cores,
		double	ram
	)

virtual

Method to find out if, right now, the compute service has at least one host with some idle number of cores and some available RAM. Note that this doesn't mean that asking for these resources right will mean immediate execution (since jobs may be pending and "ahead" in the queue, e.g., because they depend on current actions that are not using all available resources).

Parameters

num_cores	the desired number of cores
ram	the desired RAM

Returns: true if idle resources are available, false otherwise

◆ stop()

void wrench::ComputeService::stop	(	bool	send_failure_notifications,
		ComputeService::TerminationCause	termination_cause
	)

virtual

Stop the compute service.

Parameters

send_failure_notifications	whether to send job failure notifications or not
termination_cause	the cause (reason) of the service's termination

THIS IS CODE DUPLICATION FROM Service::stop(), which is not great

◆ submitCompoundJob()

virtual void wrench::ComputeService::submitCompoundJob	(	std::shared_ptr< CompoundJob >	job,
		const std::map< std::string, std::string > &	service_specific_arguments
	)

pure virtual

Method to submit a compound job to the service.

Parameters

job	The job being submitted
service_specific_arguments	the set of service-specific arguments

Implemented in wrench::HTCondorComputeService, wrench::HTCondorCentralManagerService, wrench::CloudComputeService, and wrench::BareMetalComputeService.

◆ submitJob()

void wrench::ComputeService::submitJob	(	std::shared_ptr< CompoundJob >	job,
		const std::map< std::string, std::string > &	service_specific_args = `{}`
	)

protected

Submit a job to the compute service.

Parameters

job the job

service_specific_args

arguments specific to compute services when needed:

to a BareMetalComputeService: {}
- If no entry is provided for an actionID, the service will pick on which host and with how many cores to run the task
- If a number of cores is provided (e.g., {"action1", "12"}), the service will pick the host on which to run the task
- If a hostname and a number of cores is provided (e.g., {"action1", "host1:12"}, the service will run the action on that host with the specified number of cores
to a BatchComputeService: {"-t":"<int>","-N":"<int>","-c":"<int>"[,{"-u":"<string>"}], [{actionID:[host[:num_cores]]}
- "-t": number of requested job duration in minutes
- "-N": number of requested compute hosts
- "-c": number of requested cores per compute host
- "-u": username (optional)
to a CloudComputeService: {}

Exceptions

ExecutionException
std::invalid_argument
std::runtime_error

◆ supportsCompoundJobs()

virtual bool wrench::ComputeService::supportsCompoundJobs ( )

pure virtual

Returns true if the service supports pilot jobs.

Returns: true or false

Implemented in wrench::BatchComputeService, wrench::BareMetalComputeService, wrench::CloudComputeService, wrench::HTCondorComputeService, and wrench::HTCondorCentralManagerService.

◆ supportsPilotJobs()

virtual bool wrench::ComputeService::supportsPilotJobs ( )

pure virtual

Returns true if the service supports compound jobs.

Returns: true or false

Implemented in wrench::BatchComputeService, wrench::BareMetalComputeService, wrench::CloudComputeService, wrench::HTCondorComputeService, and wrench::HTCondorCentralManagerService.

◆ supportsStandardJobs()

virtual bool wrench::ComputeService::supportsStandardJobs ( )

pure virtual

Returns true if the service supports standard jobs.

Returns: true or false

Implemented in wrench::BatchComputeService, wrench::BareMetalComputeService, wrench::CloudComputeService, wrench::HTCondorComputeService, and wrench::HTCondorCentralManagerService.

◆ terminateCompoundJob()

virtual void wrench::ComputeService::terminateCompoundJob ( std::shared_ptr< CompoundJob > job )

pure virtual

Method to terminate a compound job.

Parameters

job	the standard job

Implemented in wrench::CloudComputeService, wrench::BareMetalComputeService, wrench::HTCondorComputeService, and wrench::HTCondorCentralManagerService.

◆ terminateJob()

void wrench::ComputeService::terminateJob ( std::shared_ptr< CompoundJob > job )

Terminate a previously-submitted job (which may or may not be running yet)

Parameters

job	the job to terminate

Exceptions

std::invalid_argument
ExecutionException
std::runtime_error

◆ validateJobsUseOfScratch()

void wrench::ComputeService::validateJobsUseOfScratch ( std::map< std::string, std::string > & service_specific_args )

protectedvirtual

Method to validate that a job's use of the scratch space if ok. Throws exception if not.

Parameters

service_specific_args the job;'s service-specific arguments (useful for some services)

Reimplemented in wrench::HTCondorComputeService.

◆ validateServiceSpecificArguments()

void wrench::ComputeService::validateServiceSpecificArguments	(	std::shared_ptr< CompoundJob >	job,
		std::map< std::string, std::string > &	service_specific_args
	)

protectedvirtual

Method the validates service-specific arguments (throws std::invalid_argument if invalid)

Parameters

job	the job that's being submitted
service_specific_args	the service-specific arguments

Reimplemented in wrench::BareMetalComputeService, wrench::BatchComputeService, and wrench::HTCondorComputeService.

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

ComputeService.h
ComputeService.cpp

Public Types

Public Member Functions

Static Public Attributes

Protected Member Functions

Protected Attributes

Additional Inherited Members

Detailed Description

Constructor & Destructor Documentation

◆ ComputeService() [1/2]

◆ ComputeService() [2/2]

Member Function Documentation

◆ getCoreFlopRate()

◆ getFreeScratchSpaceSize()

◆ getHosts()

◆ getMemoryCapacity()

◆ getNumHosts()

◆ getPerHostAvailableMemoryCapacity()

◆ getPerHostNumCores()

◆ getPerHostNumIdleCores()

◆ getScratch()

◆ getScratchSharedPtr()

◆ getTotalNumCores()

◆ getTotalNumIdleCores()

◆ getTotalScratchSpaceSize()

◆ getTTL()

◆ hasScratch()

◆ isThereAtLeastOneHostWithIdleResources()

◆ stop()

◆ submitCompoundJob()

◆ submitJob()

◆ supportsCompoundJobs()

◆ supportsPilotJobs()

◆ supportsStandardJobs()

◆ terminateCompoundJob()

◆ terminateJob()

◆ validateJobsUseOfScratch()

◆ validateServiceSpecificArguments()