Classes
class	BareMetalComputeService
	A compute service that manages a set of multi-core compute hosts and provides access to their resources. More...

class	BareMetalComputeServiceMessagePayload
	Configurable message payloads for a MultiHostMulticoreComputeService. More...

class	BareMetalComputeServiceProperty
	Configurable properties for a BareMetalComputeService. More...

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

class	BatchComputeServiceMessagePayload
	Configurable message payloads for a BatchService. More...

class	BatchComputeServiceProperty
	Configurable properties for a BatchComputeService. More...

class	CloudComputeService
	A cloud-based compute service that manages a set of physical hosts and controls access to their resources by (transparently) executing jobs in VM instances. More...

class	CloudComputeServiceMessagePayload
	Configurable message payloads for a CloudService. More...

class	CloudComputeServiceProperty
	Configurable properties for a CloudComputeService. More...

class	ComputeService
	The compute service base class. More...

class	ComputeServiceMessagePayload
	Configurable message payloads for a ComputeService. More...

class	ComputeServiceProperty
	Configurable properties for a ComputeService. More...

class	EnergyMeterService
	A service that measures and records energy consumption on a set of hosts at regular time intervals. More...

class	FileRegistryService
	A file registry service (a.k.a. replica catalog) that holds a database of which files are available at which storage services. More specifically, the database holds a set of <file, storage service> entries. A WMS can add, lookup, and remove entries at will from this database. More...

class	FileRegistryServiceMessagePayload
	Configurable message payload for a FileRegistryService. More...

class	FileRegistryServiceProperty
	Configurable properties for a FileRegistryService. More...

class	HTCondorCentralManagerService
	A HTCondor central manager service implementation. More...

class	HTCondorCentralManagerServiceMessagePayload
	Configurable message payloads for an HTCondor Central Manager service. More...

class	HTCondorComputeService
	A workload management framework compute service. More...

class	HTCondorComputeServiceMessagePayload
	Configurable message payloads for an HTCondorService. More...

class	HTCondorComputeServiceProperty
	Properties for an HTCondor service. More...

class	NetworkProximityService
	A network proximity service that continuously estimates inter-host latencies and can be queried for such estimates. More...

class	NetworkProximityServiceMessagePayload
	Configurable message payloads for a NetworkProximityService. More...

class	NetworkProximityServiceProperty
	Configurable properties for a NetworkProximityService. More...

class	PegasusWorkflowParser
	A class that implement methods to read workflow files provided by the Pegasus project. More...

class	PilotJobScheduler
	A (mostly virtual) base class for implementing PilotJob scheduling algorithms to be used by a WMS. More...

class	Service
	A service that can be added to the simulation and that can be used by a WMS when executing a workflow. More...

class	ServiceMessagePayload
	Configurable message payloads for a Service. More...

class	ServiceProperty
	Configurable properties for a Service. More...

class	SimpleStorageService
	A storage service that provides direct access to some storage resources (e.g., one or more disks). An important (configurable) property of the storage service is SimpleStorageServiceProperty::BUFFER_SIZE (see documentation thereof), which defines the buffer size that the storage service uses. More specifically, when the storage service receives / sends data from / to the network, it does so in a loop over data "chunks", with pipelined network and disk I/O operations. The smaller the buffer size the more "fluid" the model, but the more time-consuming the simulation. A large buffer size, however, may lead to less realistic simulations. At the extreme, an infinite buffer size would correspond to fully sequential executions (first a network receive/send, and then a disk write/read). Setting the buffer size to "0" corresponds to a fully fluid model in which individual data chunk operations are not simulated, thus achieving both accuracy (unless one specifically wishes to study the effects of buffering) and quick simulation times. For now, setting the buffer size to "0" is not implemented. The default buffer size is 1 MiB (note that the user can always declare a disk with arbitrary bandwidth in the platform description XML). More...

class	SimpleStorageServiceMessagePayload
	Configurable message payloads for a SimpleStorageService. More...

class	SimpleStorageServiceProperty
	Configurable properties for a SimpleStorageService. More...

class	Simulation
	A class that provides basic simulation methods. Once the simulation object has been explicitly or implicitly destroyed, then any call to the WRENCH APIs has undefied behavior (due to memory being de-allocated). More...

class	SimulationOutput
	A class that contains post-mortem simulation-generated data. More...

class	SimulationTimestamp
	A time-stamped simulation event stored in SimulationOutput. More...

class	SimulationTimestampDiskRead
	A base class for simulation timestamps regarding disk reads. More...

class	SimulationTimestampDiskReadCompletion
	A simulation timestamp class for disk read completions. More...

class	SimulationTimestampDiskReadFailure
	A simulation timestamp class for disk read failure times. More...

class	SimulationTimestampDiskReadStart
	A simulation timestamp class for disk read start times. More...

class	SimulationTimestampDiskWrite
	A base class for simulation timestamps regarding disk writes. More...

class	SimulationTimestampDiskWriteCompletion
	A simulation timestamp class for disk write completions. More...

class	SimulationTimestampDiskWriteFailure
	A simulation timestamp class for disk write failure times. More...

class	SimulationTimestampDiskWriteStart
	A simulation timestamp class for disk write start times. More...

class	SimulationTimestampEnergyConsumption
	A simulation timestamp class for energy consumption. More...

class	SimulationTimestampFileCopy
	A base class for simulation timestamps regarding file copies. More...

class	SimulationTimestampFileCopyCompletion
	A simulation timestamp class for file copy completions. More...

class	SimulationTimestampFileCopyFailure
	A simulation timestamp class for file copy failure times. More...

class	SimulationTimestampFileCopyStart
	A simulation timestamp class for file copy start times. More...

class	SimulationTimestampFileRead
	A base class for simulation timestamps regarding file reads. More...

class	SimulationTimestampFileReadCompletion
	A simulation timestamp class for file read completions. More...

class	SimulationTimestampFileReadFailure
	A simulation timestamp class for file read failure times. More...

class	SimulationTimestampFileReadStart
	A simulation timestamp class for file read start times. More...

class	SimulationTimestampFileWrite
	A base class for simulation timestamps regarding file writes. More...

class	SimulationTimestampFileWriteCompletion
	A simulation timestamp class for file write completions. More...

class	SimulationTimestampFileWriteFailure
	A simulation timestamp class for file write failure times. More...

class	SimulationTimestampFileWriteStart
	A simulation timestamp class for file write start times. More...

class	SimulationTimestampPair
	A base class for simulation timestamps. More...

class	SimulationTimestampPstateSet
	A simulation timestamp class for changes in a host's pstate. More...

class	SimulationTimestampTask
	A base class for simulation timestamps regarding workflow tasks. More...

class	SimulationTimestampTaskCompletion
	A simulation timestamp class for WorkflowTask completion times. More...

class	SimulationTimestampTaskFailure
	A simulation timestamp class for WorkflowTask failure times. More...

class	SimulationTimestampTaskStart
	A simulation timestamp class for WorkflowTask start times. More...

class	SimulationTimestampTaskTermination
	A simulation timestamp class for WorkflowTask termination times. More...

class	SimulationTimestampType
	File, Source, Whoami used to be hashed as key for unordered multimap for ongoing file operations. More...

class	StandardJobScheduler
	A (mostly virtual) base class for implementing StandardJob scheduling algorithms to be used by a WMS. More...

class	StorageService
	The storage service base class. More...

class	StorageServiceMessagePayload
	Configurable message payloads for a StorageService. More...

class	StorageServiceProperty
	Configurable properties for a StorageService. More...

class	VirtualizedClusterComputeService
	A virtualized cluster-based compute service. More...

class	VirtualizedClusterComputeServiceMessagePayload
	Configurable message payloads for a VirtualizedClusterComputeService. More...

class	VirtualizedClusterComputeServiceProperty
	Configurable properties for a VirtualizedClusterComputeService. More...

class	WMS
	A workflow management system (WMS) More...

class	Workflow
	A workflow (to be executed by a WMS) More...

class	WorkflowFile
	A data file used/produced by a WorkflowTask in a Workflow. More...

class	WorkflowTask
	A computational task in a Workflow. More...

Functions
void	generateHostUtilizationGraphLayout (std::vector< WorkflowTaskExecutionInstance > &data)
	Generates graph layout for host utilization and adds that information to the JSON object. More...

std::shared_ptr< ComputeService >	getComputeService ()
	Getter. More...

std::shared_ptr< FileLocation >	getDestinationLocation ()
	Getter. More...

WorkflowFile *	getFile ()

std::string	getFunctionalityName ()
	Getter. More...

std::string	getInvalidPath ()
	Get the invalid path. More...

WorkflowJob *	getJob ()
	Get the executor's job. More...

std::shared_ptr< FileLocation >	getLocation ()
	Getter. More...

std::string	getMailbox ()
	Returns the mailbox name on which the error occurred. More...

std::shared_ptr< Service >	getService ()

std::shared_ptr< FileLocation >	getSourceLocation ()
	Getter. More...

std::shared_ptr< StorageService >	getStorageService ()
	Get the location's storage service. More...

bool	isPointOnSegment (std::pair< unsigned long long, unsigned long long > segment, unsigned long long point)
	Determines if a point lies on a line segment. More...

bool	isSegmentOverlappingXAxis (std::pair< unsigned long long, unsigned long long > segment1, std::pair< unsigned long long, unsigned long long > segment2)
	Determines if two line segments overlap along the x-axis and allows for a slight overlap. More...

bool	isSegmentOverlappingYAxis (std::pair< unsigned long long, unsigned long long > segment1, std::pair< unsigned long long, unsigned long long > segment2)
	Determines if two line segments overlap along the y-axis using exact values. More...

bool	isTimeout ()
	Returns whether the network error was a timeout. More...

bool	operator== (const BatchJobSet &left, const BatchJobSet &right)
	Overloaded operator. More...

static std::tuple< std::string, unsigned long >	parseResourceSpec (std::string spec)
	Helper static method to parse resource specifications to the <cores,ram> format. More...

bool	searchForLayout (std::vector< WorkflowTaskExecutionInstance > &data, std::size_t index)
	Searches for a possible host utilization gantt chart layout and updates the data to include what vertical position to plot each rectangle. More...

void	to_json (nlohmann::json &j, const WorkflowTaskExecutionInstance &w)
	Function called by the nlohmann::json constructor when a WorkflowTaskExecutionInstance is passed in as a parameter. This returns the JSON representation of a WorkflowTaskExecutionInstance. The name of this function is important and should not be changed as it is what nlohmann expects (hardcoded in there). More...

std::string	toString ()=0
	Return an error message that describes the failure cause (to be overridden) More...

bool	whileReceiving ()
	Returns whether the network error occurred while receiving. More...

bool	whileSending ()
	Returns whether the network error occurred while sending. More...

Detailed Description

This program is free software: you can redistribute it and/or modify it under the terms of the GNU General Public License as published by the Free Software Foundation, either version 3 of the License, or (at your option) any later version.

Function Documentation

◆ generateHostUtilizationGraphLayout()

void wrench::generateHostUtilizationGraphLayout ( std::vector< WorkflowTaskExecutionInstance > & data )

Generates graph layout for host utilization and adds that information to the JSON object.

Searches for a possible gantt chart layout to represent host utilization. If a layout is found (no tasks overlap), then information about where to plot what is added to the JSON object. Note that this is a possible layout and does not reflect what task ran on what core specifically. For example, we may hav a task that was allocated 2-cores on a idle 4-core host. The task, when plotted on the gantt chart may end up in 1 of 3 positions (using cores 0 and 1, 1 and 2, or 2 and 3).

Parameters

data	JSON workflow execution data

Exceptions

std::runtime_error

◆ getComputeService()

std::shared_ptr< ComputeService > wrench::getComputeService ( )

Getter.

Get the compute service provided by the (running) pilot job.

Returns: the compute service; a compute service

◆ getDestinationLocation()

std::shared_ptr< FileLocation > wrench::FileAlreadyBeingCopied::getDestinationLocation ( )

Getter.

Returns: the source location

◆ getFunctionalityName()

std::string wrench::FunctionalityNotAvailable::getFunctionalityName ( )

Getter.

Returns: the functionality name

◆ getInvalidPath()

std::string wrench::InvalidDirectoryPath::getInvalidPath ( )

Get the invalid path.

Returns: a path

◆ getJob()

WorkflowJob * wrench::getJob ( )

Get the executor's job.

Get the task's containing job.

Getter.

Retrieve the job the WorkunitExecutor is working for.

Retrieve the standard job this workunit belongs to.

Returns: a standard job; a job; the job; job: the job

◆ getLocation()

std::shared_ptr< FileLocation > wrench::FileNotFound::getLocation ( )

Getter.

Returns: the storage service

◆ getMailbox()

std::string wrench::NetworkError::getMailbox ( )

Returns the mailbox name on which the error occurred.

Returns: the mailbox name

◆ getSourceLocation()

std::shared_ptr< FileLocation > wrench::FileAlreadyBeingCopied::getSourceLocation ( )

Getter.

Returns: the source location

◆ getStorageService()

std::shared_ptr< StorageService > wrench::getStorageService ( )

Get the location's storage service.

Getter.

Get the storage service at which the path was invalid.

Returns: a storage service; the storage service

◆ isPointOnSegment()

bool wrench::isPointOnSegment	(	std::pair< unsigned long long, unsigned long long >	segment,
		unsigned long long	point
	)

Determines if a point lies on a line segment.

Parameters

segment	pair of start and end points that make up the line segment
point	point on a 1D plane

Returns: bool

◆ isSegmentOverlappingXAxis()

bool wrench::isSegmentOverlappingXAxis	(	std::pair< unsigned long long, unsigned long long >	segment1,
		std::pair< unsigned long long, unsigned long long >	segment2
	)

Determines if two line segments overlap along the x-axis and allows for a slight overlap.

Parameters

segment1	first segment
segment2	second segment

Returns: bool

◆ isSegmentOverlappingYAxis()

bool wrench::isSegmentOverlappingYAxis	(	std::pair< unsigned long long, unsigned long long >	segment1,
		std::pair< unsigned long long, unsigned long long >	segment2
	)

Determines if two line segments overlap along the y-axis using exact values.

Parameters

segment1	first segment
segment2	second segment

Returns: bool

◆ isTimeout()

bool wrench::NetworkError::isTimeout ( )

Returns whether the network error was a timeout.

Returns: true or false

◆ operator==()

bool wrench::operator==	(	const BatchJobSet &	left,
		const BatchJobSet &	right
	)

Overloaded operator.

Parameters

left	left-hand side
right	right-hand side

Returns

◆ parseResourceSpec()

static std::tuple<std::string, unsigned long> wrench::parseResourceSpec ( std::string spec )

static

Helper static method to parse resource specifications to the <cores,ram> format.

Parameters

spec	specification string

Returns: a <cores, ram> tuple

Exceptions

std::invalid_argument

◆ searchForLayout()

bool wrench::searchForLayout	(	std::vector< WorkflowTaskExecutionInstance > &	data,
		std::size_t	index
	)

Searches for a possible host utilization gantt chart layout and updates the data to include what vertical position to plot each rectangle.

Recursive backtracking search for a valid gantt chart layout. This algorithm looks for a vertical position to place each task execution event such that it doesn't overlap with any other task.

Parameters

data	JSON workflow execution data
index	the index of the workflow execution data up to where we would like to check for a valid layout

Returns: bool

◆ to_json()

void wrench::to_json	(	nlohmann::json &	j,
		const WorkflowTaskExecutionInstance &	w
	)

Function called by the nlohmann::json constructor when a WorkflowTaskExecutionInstance is passed in as a parameter. This returns the JSON representation of a WorkflowTaskExecutionInstance. The name of this function is important and should not be changed as it is what nlohmann expects (hardcoded in there).

Parameters

j	reference to a JSON object
w	reference to a WorkflowTaskExecutionInstance

◆ toString()

std::string wrench::toString ( )

override

Return an error message that describes the failure cause (to be overridden)

Get the human-readable failure message.

Give a <ss name>:<mount point>:<dir>" string for the location.

Returns: an error message; A string; the message

◆ whileReceiving()

bool wrench::NetworkError::whileReceiving ( )

Returns whether the network error occurred while receiving.

Returns: true or false

◆ whileSending()

bool wrench::NetworkError::whileSending ( )

Returns whether the network error occurred while sending.

Returns: true or false

Classes

Functions

Detailed Description

Function Documentation

◆ generateHostUtilizationGraphLayout()

◆ getComputeService()

◆ getDestinationLocation()

◆ getFunctionalityName()

◆ getInvalidPath()

◆ getJob()

◆ getLocation()

◆ getMailbox()

◆ getSourceLocation()

◆ getStorageService()

◆ isPointOnSegment()

◆ isSegmentOverlappingXAxis()

◆ isSegmentOverlappingYAxis()

◆ isTimeout()

◆ operator==()

◆ parseResourceSpec()

◆ searchForLayout()

◆ to_json()

◆ toString()

◆ whileReceiving()

◆ whileSending()