ScienceDirect® Home Skip Main Navigation Links
You have guest access to ScienceDirect. Find out more.
 
Home
Browse
My Settings
Alerts
Help
 Quick Search
 Search tips (Opens new window)
    Clear all fields    
Computer Networks
Volume 51, Issue 17, 5 December 2007, Pages 4816-4835
 
Font Size: Decrease Font Size  Increase Font Size
 Abstract - selected
Article
Purchase PDF (668 K)

  E-mail Article   
  Add to my Quick Links   
Bookmark and share in 2collab (opens in new window)
Request permission to reuse this article
  Cited By in Scopus (0)
 
 
 
Related Articles in ScienceDirect
View More Related Articles
 
View Record in Scopus
 
doi:10.1016/j.comnet.2007.07.006    How to Cite or Link Using DOI (Opens New Window)
Copyright © 2007 Elsevier B.V. All rights reserved.

Statistical models for Gnutella signaling traffic

Dragos IlieCorresponding Author Contact Information, a, E-mail The Corresponding Author and Adrian Popescua, E-mail The Corresponding Author

aSchool of Engineering, Department of Telecommunication Systems, Blekinge Institute of Technology (BTH), 371 39 Karlskrona, Sweden

Received 11 January 2007; 
revised 28 June 2007; 
accepted 22 July 2007. 
Responsible Editor: Minglu Li. 
Available online 6 August 2007.

Purchase the full-text article



References and further reading may be available for this article. To view references and further reading you must purchase this article.

Abstract

The paper is focused on signaling traffic between Gnutella peers that implement the latest Gnutella protocol specifications (v0.6). In particular, we provide analytically tractable statistical models at session level, message level and IP datagram level for traffic crossing a Gnutella ultrapeer at Blekinge Institute of Technology (BTH) in Karlskrona, Sweden. To the best of our knowledge this is the first work that provides Gnutella v0.6 statistical models at this level of detail. These models can be implemented straightforward in network simulators such as ns2 and OmNet++.

The results show that incoming requests to open a session follow a Poisson distribution. Incoming Gnutella messages across all established sessions can be described by a compound Poisson distribution. Mixture distribution models for message transfer rates include a heavy-tailed component.

Keywords: Gnutella; Signaling traffic; Statistical models; Traffic measurements

Article Outline

1. Introduction
2. Related work
3. The Gnutella protocol
3.1. Bootstrap
3.2. Peer connections
3.3. Gnutella message headers
3.4. Peer discovery
3.5. Resource discovery
3.6. Query routing protocol (QRP)
3.7. Content distribution
3.8. Horizon size estimation protocol (HSEP)
4. Measurement infrastructure
5. Methodology for statistical modeling
5.1. Model construction
5.2. Fitness assessment
5.3. Mixture distributions
5.4. Methodology review
5.5. Numerical software and methods
6. Statistical models
6.1. Ultrapeer settings and packet-trace statistics
6.2. Session characteristics
6.2.1. Session interarrival and interdeparture times
6.2.2. Session size and duration
6.3. Message characteristics
6.4. Transfer rate characteristics
6.5. Traffic characteristics at IP layer
7. Conclusions
References
Vitae













Computer Networks
Volume 51, Issue 17, 5 December 2007, Pages 4816-4835
 
Home
Browse
My Settings
Alerts
Help
Elsevier.com (Opens new window)
About ScienceDirect  |  Contact Us  |  Information for Advertisers  |  Terms & Conditions  |  Privacy Policy
Copyright © 2008 Elsevier B.V. All rights reserved. ScienceDirect® is a registered trademark of Elsevier B.V.