Clustering players for load balancing in virtual worlds

Authors:
Simon Rieche;Klaus Wehrle;Marc Fouquet;Heiko Niedermayer;Timo Teifel;Georg Carle
Affiliations:
Distributed Systems Group, RWTH Aachen University, Ahornstrasse 55, 52074 Aachen, Germany.;Distributed Systems Group, RWTH Aachen University, Ahornstrasse 55, 52074 Aachen, Germany.;Computer Networks and Internet Group, University of Tubingen, Sand 13, 72076 Tubingen, Germany.;Computer Networks and Internet Group, University of Tubingen, Sand 13, 72076 Tubingen, Germany.;Computer Networks and Internet Group, University of Tubingen, Sand 13, 72076 Tubingen, Germany.;Network Architectures and Services Group, Technical University of Munich, Boltzmannstrasse 3, 85748 Munchen, Germany
Venue:
International Journal of Advanced Media and Communication
Year:
2008

Citing 4
Cited 1

A scalable content-addressable network

Proceedings of the 2001 conference on Applications, technologies, architectures, and protocols for computer communications
Low latency and cheat-proof event ordering for peer-to-peer games

NOSSDAV '04 Proceedings of the 14th international workshop on Network and operating systems support for digital audio and video
Simple efficient load balancing algorithms for peer-to-peer systems

Proceedings of the sixteenth annual ACM symposium on Parallelism in algorithms and architectures
Zoned federation of game servers: a peer-to-peer approach to scalable multi-player online games

Proceedings of 3rd ACM SIGCOMM workshop on Network and system support for games

Flexible load distribution for hybrid distributed virtual environments

Future Generation Computer Systems

Quantified Score

Hi-index	0.00

Visualization

Abstract

In current Massively Multiplayer Online Games (MMOGs) the distribution of load is not distributed dynamically. But players tend to concentrate in certain regions of the game world and cause overload conditions. Therefore we propose the use of structured Peer-to-Peer technology for the server infrastructure of the MMOGs to improve the reliability and scalability. Previous work segmented the game work into rectangular areas but often split a group of players to different servers. This work presents a cluster-based P2P approach, which is able to dynamically adapt to the current state of the game and handle uneven distributions of players. We show through simulation, also with traces from real online games, that the cluster-based approach performs better than the previous P2P-based system.