Churn generator producing diurnal usage patterns


This tool [4] can be used to generate tracefiles for churn in P2P systems that exhibits diurnal usage patterns similar to those observed in [1] and [2]. The churn generator comes with two modes one based on the session data from Steiner [1] producing churn with few but longer sessions and one mode based on data from Schneider [2] which produces many more but shorter sessions per day and node.

Dependencies

Python 2.7.x
Numpy 1.7.x
SciPy 0.12.x

After the installation, tests can be executed using python'(m'unittest'discover

Usage

Execute  python'steiner.py'(h or python'schneider.py'(h to get detailed usage information.


The following example creates a trace with 1000 nodes starting on second 43200 (Monday 12:00 UTC+0) for a total duration of 604800 seconds (seven days) using a timezone distribution from Pingdom[3]:

python steiner.py -n 1000 -s 43200 -d 604800 world

License

The code is open-source and licensed under GPLv3.

Copyright (C) 2013 Niklas Lochschmidt nlochschmidt(a-t)gmail.com

 

1. M. Steiner, T. En-Najjary, and E. Biersack, “Long term study of peer behavior in the KAD DHT,” IEEE/ACM Transactionson Networking, vol. 17, no. 5, pp. 1371–1384, 2009.


2. F. Schneider, A. Feldmann, B. Krishnamurthy, and W. Willinger, “Understanding online social network usage from a network perspective,” in Proceedings of the 9th ACM SIGCOMM conference on Internet measurement conference, 2009, pp. 35–48.


3. http://royal.pingdom.com/2013/02/12/internet-users-time-zone/

 

4. Download link: Churn generator

A A A | Drucken Print | Impressum Impressum | Sitemap Sitemap | Suche Search | Kontakt Contact | Webseitenanalyse: Mehr Informationen
zum Seitenanfangzum Seitenanfang