omnetpypy.simulation

This module contains a set of global configuration variables for the simulation, such as the engine to use, and a set of functions to make the configuration more user-friendly.

class omnetpypy.simulation.Experiment(config_file)[source]

Bases: object

This class is used to run a set of independent simulations using the same YAML configuration files. An instance of this class takes care of reading the configuration file, parsing it, running the simulations, and dumping the recorded metrics in the output files.

Parameters:

config_filestr

The path to the configuration file, absolute or relative to the working directory. The configuration file is a YAML file with a rigid structure (see config attribute).

Examples

>>> exp = Experiment(config_file="config.yaml")
>>> exp.run_simulations()

Attributes:

seed_setslist of lists of int

The seed sets for each repetition. Each repetition has a seed set of length “rngs_per_rep” from the configuration file.

configdict

The configuration dictionary, directly parsed from the YAML configuration file. The dictionary has the following keys:

• engine (str), optional:

  The simulation engine to use. Only “simpy” is supported at the moment. Defaults to “simpy”.

• repetitions (int), optional:

  The number of independent repetitions to run. Defaults to 1.

• metrics (list of FutureMetric), optional:

  The list of metrics to collect. Defaults to an empty list.

• yaml_directory (str), optional:

  The path to the directory containing the yaml files to set the simulation. Relative to the configuration file directory. Defaults to “./” .

• simulate_until (float or None), optional:

  The time until which the simulation will run. If None, simulation runs as long as there are events. Defaults to None.

• time_unit (str), optional:

  The time unit for the simulation. Defaults to “us” (microseconds).

• log_level (int or str), optional:

  The log level for the simulation. See set_log_level(). Defaults to logging.INFO

• output_dir (str or None), optional:

  The output directory for the simulation, relative to the configuration file directory. Defaults to None, in which case no data will be stored.

simulations_paramslist of tuples

The list of simulation parameters for each repetition. Each element of the list is a tuple with the following parameters:

• engine (str):

  The simulation engine to use.

• seed_set (list of int):

  The seed set for the repetition.

• repetition (int):

  The repetition index.

• metrics (list of FutureMetric):

  The list of metrics to collect.

• yaml_directory (str):

  The path to the directory containing the yaml files to set the simulation.

• until (float or None):

  The time until which the simulation will run. If None, simulation runs as long as there are events.

• log_level (int or str):

  The log level for the simulation. See set_log_level().

• time_unit (str):

  The time unit for the simulation.

• output_dir (str or None):

  The output directory for the simulation.

• global_params (dict):

  A dictionary of global parameters. These parameters are available to all the entities in the simulation.

output_dirstr or None

The output directory for the simulation. If None, no data is stored.

run_simulations()[source]

Run the simulations.

class omnetpypy.simulation.Simulation(engine, seed_set, repetition, metrics, yaml_directory, until, log_level, time_unit, output_dir, global_params)[source]

Bases: object

This class is used to run a single simulation using the specified configuration. It takes care of reading the running the simulation and returning the recorded metrics. Users should not instantiate this class directly, but use the Experiment to manage multiple simulations.

All parameters are also stored as attributes.

Parameters:

enginestr

The simulation engine to use. Only “simpy” is implemented at the moment.

seed_setlist of int

The seed set for the repetition.

repetitionint

The repetition index.

metricslist of FutureMetric

The list of metrics to collect.

yaml_directorystr

The path of the directory that contains all the yaml files.

untilint or float

The time until which the simulation will run.

log_levelstr

The log level for the simulation. See log_to_console().

time_unitstr

The time unit for the simulation. See time_unit_factor().

output_dirstr

The output directory for the simulation.

global_paramsdict

A dictionary of global parameters. These parameters are available to all the entities in the simulation.

Raises:

NotImplementedError

If the passed engine is not implemented.

Attributes:

rngMultiRandom

The random number generator for this simulation, already initialized. It is made available to all the entities in the simulation through the Connector.

connectorConnector

The connector to the simulation engine. Its subclass depends on the chosen engine.

networkCompoundModule

The network to simulate, parsed from the YAML file “network.yaml”.

start()[source]

start the simulation and collect the metrics.

Returns:

dict

A dictionary of dictionaries with the collected metrics.

time()[source]

Return the current simulation time.

Returns:

int or float

The current simulation time.

property time_unit_factor

The factor to convert the simulation time unit to seconds.

Returns:

int or float

The factor to convert the time unit to seconds.

See also

time_unit_factor()