wrench::BatchService Class Reference

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

#include <BatchService.h>

Inheritance diagram for wrench::BatchService:
wrench::ComputeService wrench::Service

Public Member Functions

 BatchService (std::string &hostname, std::vector< std::string > compute_hosts, double scratch_space_size, std::map< std::string, std::string > property_list={}, std::map< std::string, std::string > messagepayload_list={})
 Constructor. More...
 
std::map< std::string, double > getStartTimeEstimates (std::set< std::tuple< std::string, unsigned int, unsigned int, double >>)
 Retrieve start time estimates for a set of job configurations. More...
 
- Public Member Functions inherited from wrench::ComputeService
std::vector< 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< double > getMemoryCapacity ()
 Get the RAM capacities for each of the compute service's hosts. More...
 
std::vector< unsigned long > getNumCores ()
 Get core counts for each of the compute service's host. More...
 
unsigned long getNumHosts ()
 Get the number of hosts that the compute service manages. More...
 
std::vector< unsigned long > getNumIdleCores ()
 Get idle core counts for each of the compute service's host. More...
 
StorageServicegetScratch ()
 Get the compute service's scratch storage space. More...
 
std::shared_ptr< StorageServicegetScratchSharedPtr ()
 Get a shared pointer to the compute service's scratch storage space. 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...
 
void stop () override
 Stop the compute service - must be called by the stop() method of derived classes.
 
void submitJob (WorkflowJob *job, std::map< std::string, std::string >={})
 Submit a job to the compute service. More...
 
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 (WorkflowJob *job)
 Terminate a previously-submitted job (which may or may not be running yet) More...
 
- Public Member Functions inherited from wrench::Service
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...
 
bool isUp ()
 Returns true if the service is UP, false otherwise. More...
 
void setNetworkTimeoutValue (double value)
 Sets the service's network timeout value. More...
 
void start (std::shared_ptr< Service > this_service, bool daemonize=false)
 Start the service. More...
 

Additional Inherited Members

- Public Types inherited from wrench::Service
enum  State { UP, DOWN }
 Service states. More...
 
- 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.
 
static StorageServiceSCRATCH = (StorageService *)ULONG_MAX
 A convenient constant to mean "the scratch storage space" of a ComputeService. This is used to move data to a ComputeService's scratch storage space.
 

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 partitioning among jobs on 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 contended scenarios is thus, for now, not realistic as there is no simulation of the effects of memory sharing (e.g., swapping).

Constructor & Destructor Documentation

wrench::BatchService::BatchService ( std::string &  hostname,
std::vector< std::string >  compute_hosts,
double  scratch_space_size,
std::map< std::string, std::string >  property_list = {},
std::map< std::string, std::string >  messagepayload_list = {} 
)

Constructor.

Parameters
hostnamethe hostname on which to start the service
compute_hoststhe 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_sizethe size for the scratch storage space for the service (0 means "no scratch space")
property_lista property list that specifies BatchServiceProperty values ({} means "use all defaults")
messagepayload_lista message payload list that specifies BatchServiceMessagePayload values ({} means "use all defaults")

Member Function Documentation

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

Retrieve start time estimates for a set of job configurations.

Parameters
set_of_jobsthe 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 int)
  • a number of cores per host (unsigned int)
  • 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:
  • /home/wrench/wrench/include/wrench/services/compute/batch/BatchService.h
  • /home/wrench/wrench/src/wrench/services/compute/batch/BatchService.cpp