trafpy.benchmarker.versions package

Subpackages

• trafpy.benchmarker.versions.benchmark_v001 package
  • Submodules
  • trafpy.benchmarker.versions.benchmark_v001.benchmark_plot_script module
  • trafpy.benchmarker.versions.benchmark_v001.config module
  • trafpy.benchmarker.versions.benchmark_v001.distribution_generator module
  • trafpy.benchmarker.versions.benchmark_v001.old_distribution_generator module
  • Module contents

Submodules

trafpy.benchmarker.versions.benchmark module

class trafpy.benchmarker.versions.benchmark.Benchmark(benchmark_name, benchmark_version='v001', load_prev_dists=True, jobcentric=False)

Bases: ABC

get_dist_and_path(dist_name)

Gets distribution data and corresponding path.

If distribution data does not exist, will return dist=None and the path will be where the dist data should be saved if it is generated.

abstract get_flow_size_dist(dist_name='flow_size_dist')

Loads previously saved flow size dist (if it exists).

This is an abstract method and therefore must be defined by any child class.

Parameters

dist_name (str) – Name of distribution (determines path to search for previously saved distribution).

abstract get_interarrival_time_dist(dist_name='interarrival_time_dist')

Loads previously saved interarrival time dist (if it exists).

This is an abstract method and therefore must be defined by any child class.

Parameters

dist_name (str) – Name of distribution (determines path to search for previously saved distribution).

abstract get_node_dist(eps, racks_dict, dist_name='node_dist')

Loads previously saved node dist (if it exists).

This is an abstract method and therefore must be defined by any child class.

Parameters

• eps (list) – List of network end points.

• racks_dict (dict) – Dict mapping racks to the corresponding end points contained within each rack.

• dist_name (str) – Name of distribution (determines path to search for previously saved distribution).

get_num_ops_dist(dist_name='num_ops_dist')

Loads previously saved number of operations dist (if it exists).

This method only needs to be defined by a child class if jobcentric=True, since flow-centric data have no notion of ‘number of operations’ (in relation to job DAGs).

Parameters

dist_name (str) – Name of distribution (determines path to search for previously saved distribution).

load_dist(benchmark_name, dist_name)

Loads previously saved distribution data for given benchmark.

save_dist(dist_data, dist_name)

Saves distribution data for a given benchmark.

trafpy.benchmarker.versions.benchmark_importer module

class trafpy.benchmarker.versions.benchmark_importer.BenchmarkImporter(benchmark_version='v001', load_prev_dists=True)

Bases: object

get_benchmark_dists(benchmark_name, eps, racks_dict=None)

Retrieves pre-defined TrafPy benchmark distributions for a network.

Parameters

• benchmark_name (str) – Name of benchmark (e.g. ‘university’).

• eps (list) – List of end points/machines/leaf nodes in network.

• racks_dict (dict) – Mapping of which end points are in which racks. Keys are rack ids, values are list of end points. If None, assume there is not clustering/rack system in the network where have different end points in different clusters/racks.

trafpy.benchmarker.versions.old_benchmark_importer module

class trafpy.benchmarker.versions.old_benchmark_importer.BenchmarkImporter(benchmark_version, load_prev_dists=True)

Bases: object

get_benchmark_dists(benchmark, eps, racks_dict=None)

Retrieves pre-defined TrafPy benchmark distributions for a network.

Parameters

• benchmark (str) – Name of benchmark (e.g. ‘university’).

• eps (list) – List of end points/machines/leaf nodes in network.

• racks_dict (dict) – Mapping of which end points are in which racks. Keys are rack ids, values are list of end points. If None, assume there is not clustering/rack system in the network where have different end points in different clusters/racks.

Module contents