Please use this identifier to cite or link to this item: http://hdl.handle.net/1959.3/2535

Download PDF (Accepted manuscript) (Adobe Acrobat PDF, -1 bytes)

Title

Dissecting server-discovery traffic patterns generated by multiplayer first person shooter games

Author(s)

Zander, Sebastian;  Kennedy, David;  Armitage, Grenville

Abstract

We study the 'background traffic' resulting from tens of thousands of networked first person shooter (FPS) clients searching for servers on which to play. Networked, multiplayer games utilise the network in two distinct ways. Game play is typically built around a client-server communication model, and the resulting traffic patterns have been well studied to date. However, the discovery of available game servers is itself a client-server process. Operational game servers register themselves with well-known 'master servers', which are then queried by game clients looking for available servers. Game clients then probe the servers and retrieve information such as game type, number of other players, currently active map, and latency (ping time). We instrumented two active and public "Wolfenstein: Enemy Territory" servers over 20 weeks, developed a simple method to differentiate client probes from game-play traffic, and then characterized and contrasted the time-of-day, geographical distributions and traffic characteristics of both traffic types. We find that a significant amount of a server's traffic is probe traffic and the geographical origins are very different for both types of traffic. We propose techniques to improve server location and to decrease the amount of probe traffic.

Publication type

Conference paper

Research centre

Swinburne University of Technology. Faculty of Information and Communication Technologies. Centre for Advanced Internet Architectures

Source

Proceedings of the 4th Workshop on Network and System Support for Games (NETGAMES '05), Hawthorne, New York, United States, 10-11 October 2005, pp. 1-12

Publication year

2005

Keyword(s)

Game traffic analysis;  Geographic distribution;  Server location

Publisher

ACM

ISBN

1595931570

Publisher URL

http://dx.doi.org/10.1145/1103599.1103620

Copyright

Copyright © ACM, 2005. This is the author's version of the work. It is posted here by permission of ACM for your personal use. Not for redistribution. The definitive version was published in Proceedings of NETGAMES '05 (2005) http://doi.acm.org/10.1145/1103599.1103620

Full text

Peer reviewed