Zoneless load balancing for massively multiplayer online games

Bamutange, B., Ramsurrun, Visham, Seeam, Amar ORCID logoORCID:, Katsina, P. and Anantwar, S. (2020) Zoneless load balancing for massively multiplayer online games. In: 2020 3rd International Conference on Emerging Trends in Electrical, Electronic and Communications Engineering (ELECOM), 25-27 November 2020, Mauritius. . [Conference or Workshop Item] (doi:10.1109/ELECOM49001.2020.9296989)


Online gaming has greatly improved today especially among Massively Multiplayer Online Games which are able to handle an ever-growing number of simultaneous players participating within a single virtual game space. Typical load balancing techniques involve partitioning the game world into segments (zones), with each zone run by a group of servers and utilize standard Client/Server architectures. Zoning, however, presents some challenges that remain unresolved such as the negative effect of player crowding on system performance. Standard Client/Server architectures in MMOGs also experience limitations related to latency, reliability, and scalability. This paper proposes a proof of concept zoneless load balancing technique which is based on a hybrid architecture that consists of a distributed backend to minimize server load differences. To eliminate the need for zoning, a single Master node is implemented to coordinate cluster connections through TCP/IP and Slaves are divided into "Gateways" and "Workers". The players will connect to Gateway nodes which handle player characters through Client/Server network communications. Worker nodes will handle the non-player components of the virtual world such as weather, terrain and non-player characters (NPCs) and each component can be assigned its own Worker node(s) to run on. All nodes in the cluster will communicate with each other via incoming and outgoing host/port addresses. All game objects will ultimately render onto the same virtual space and appear to the player as a single game world.

Item Type: Conference or Workshop Item (Paper)
Research Areas: A. > School of Science and Technology > Computer Science
Item ID: 36101
Depositing User: Amar Kumar Seeam
Date Deposited: 05 Oct 2022 11:07
Last Modified: 05 Oct 2022 11:07

Actions (login required)

View Item View Item


Activity Overview
6 month trend
6 month trend

Additional statistics are available via IRStats2.