Example: 'Vary a peak demand set according to an activity function depending on the time and timezone'

Brief description

This generator modifies the peak offered-traffic volume per demand according to an activity function, according to the current time and the node timezone

Algorithm description table

Algorithm inputs

Requires a demand set within the netPlan object.

Algorithm parameters:

  • defaultTimezone. Default timezone with respect to UTC (in range [-12, 12]). Default: 0
  • sigma. Standard deviation from normalized peak traffic. Default: 0.05
  • threshold. Maximum variation of the random noise. Default: 0.25
  • randomSeed. Seed for the random generator (-1 means random). Default: -1
Algorithm outputsThe demand set scaled according to the activity function, with a small random noise
Required librariesNone
KeywordsNone
AuthorsPablo Pavón Mariño, José Luis Izquierdo Zaragoza
DateMarch 2013
CodeTVSim_EG_activityFunction.java

Detailed description

This algorithm modifies the peak offered-traffic volume per demand according to an activity function (see [1]) with respect to the timezone. The activity of a node-pair is assumed to be equal to the average of the respective activity function of each node. In its turn, in [1] it is the product.

The timezone of each node is represented as the time offset respecting to Coordinated Universal Time (UTC) and it is between -12 and 12.

Early-morning and evening traffic is assumed to be 30% of the peak values, and on weekends the peak value is assumed to be 50% of that on working days. A Gaussian randomness is added to each demand with a typical deviation of 5% of its traffic.

[1] J. Milbrandt, M. Menth, S. Kopf, "Adaptive Bandwidth Allocation: Impact of Traffic Demand Models for Wide Area Networks," University of Würzburg, Germany, Report No. 363, June 2005