A batch-scheduled compute service that manages a set of compute hosts and controls access to their resource via a batch queue. More...

#include <BatchComputeService.h>

Inheritance diagram for wrench::BatchComputeService:

Public Member Functions
	BatchComputeService (const std::string &hostname, std::vector< std::string > compute_hosts, std::string scratch_space_mount_point, std::map< std::string, std::string > property_list={}, std::map< std::string, double > messagepayload_list={})
	Constructor. More...

std::vector< std::tuple< std::string, std::string, int, int, int, double, double > >	getQueue ()
	Gets the state of the batch queue. More...

std::map< std::string, double >	getStartTimeEstimates (std::set< std::tuple< std::string, unsigned long, unsigned long, double >> resources)
	Retrieve start time estimates for a set of job configurations. More...

Public Member Functions inherited from wrench::ComputeService
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 >	getScratchSharedPtr ()
	Get a shared pointer to 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. 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...

bool	hasScratch ()
	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. More...

void	stop () override
	Stop the compute service - must be called by the stop() method of derived classes.

bool	supportsPilotJobs ()
	Get whether the compute service supports pilot jobs or not. More...

bool	supportsStandardJobs ()
	Get whether the compute service supports standard jobs or not. More...

void	terminateJob (std::shared_ptr< WorkflowJob > 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...

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

bool	getPropertyValueAsBoolean (std::string)
	Get a property of the Service as a boolean. More...

double	getPropertyValueAsDouble (std::string)
	Get a property of the Service as a double. More...

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

unsigned long	getPropertyValueAsUnsignedLong (std::string)
	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	start (std::shared_ptr< Service > this_service, bool daemonize, bool auto_restart)
	Start the service. More...

void	suspend ()
	Suspend the service.

Additional Inherited Members
Static Public Attributes inherited from wrench::ComputeService
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.

Detailed Description

A batch-scheduled compute service that manages a set of compute hosts and controls access to their resource via a batch queue.

In the current implementation of this service, like for many of its real-world counterparts, memory_manager_service partitioning among jobs onq the same host is not handled. When multiple jobs share hosts, which can happen when jobs require only a few cores per host and can thus be co-located on the same hosts in a non-exclusive fashion, each job simply runs as if it had access to the full RAM of each compute host it is scheduled on. The simulation of these memory_manager_service contended scenarios is thus, for now, not realistic as there is no simulation of the effects of memory_manager_service sharing (e.g., swapping).

Constructor & Destructor Documentation

◆ BatchComputeService()

wrench::BatchComputeService::BatchComputeService	(	const std::string &	hostname,
		std::vector< std::string >	compute_hosts,
		std::string	scratch_space_mount_point,
		std::map< std::string, std::string >	property_list = `{}`,
		std::map< std::string, double >	messagepayload_list = `{}`
	)

Constructor.

Parameters

hostname	the hostname on which to start the service
compute_hosts	the list of names of the available compute hosts the hosts must be homogeneous (speed, number of cores, and RAM size) all cores are usable by the batch service on each host all RAM is usable by the batch service on each host
scratch_space_mount_point	the mount point of the scratch storage space for the service ("" means "no scratch space")
property_list	a property list that specifies BatchComputeServiceProperty values ({} means "use all defaults")
messagepayload_list	a message payload list that specifies BatchComputeServiceMessagePayload values ({} means "use all defaults")

Member Function Documentation

◆ getQueue()

std::vector< std::tuple< std::string, std::string, int, int, int, double, double > > wrench::BatchComputeService::getQueue ( )

Gets the state of the batch queue.

Returns

A vector of tuples:

std::string: username
string: job name
int: num hosts
int: num cores per host
int: time in seconds
double: submit time
double: start time (-1.0 if not started yet)

◆ getStartTimeEstimates()

std::map< std::string, double > wrench::BatchComputeService::getStartTimeEstimates ( std::set< std::tuple< std::string, unsigned long, unsigned long, double >> set_of_jobs )

Retrieve start time estimates for a set of job configurations.

Parameters

set_of_jobs

the set of job configurations, each of them with an id. Each configuration is a tuple as follows:

a configuration id (std::string)
a number of hosts (unsigned long)
a number of cores per host (unsigned long)
a duration in seconds (double)

Returns: start date predictions in seconds (as a map of ids). A prediction that's negative means that the job configuration can not run on the service (e.g., not enough hosts, not enough cores per host)

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

BatchComputeService.h
BatchComputeService.cpp