The Community for Technology Leaders
Green Image
Issue No. 06 - June (2010 vol. 21)
ISSN: 1045-9219
pp: 765-777
Xueyan Tang , Nanyang Technological University, Singapore
Suiping Zhou , Nanyang Technological University, Singapore
The fundamental goal of distributed virtual environments (DVEs) is to create a common and consistent presentation of the virtual world among a set of computers interconnected by a network. This paper investigates update scheduling algorithms to make efficient use of network capacity and improve consistency in DVEs. Our approach is to schedule state updates according to their potential impacts on consistency. In DVEs, the perceptions of participants are affected by both the spatial magnitude and temporal duration of inconsistency in the virtual world. Using the metric of time-space inconsistency, we analytically derive the optimal update schedules for minimizing the impact of inconsistency. Based on the analysis, we propose a number of scheduling algorithms that integrate spatial and temporal factors. These algorithms also take into consideration the effect of network delays. The algorithms can be used on top of many existing mechanisms such as dead reckoning. Experimental results show that our proposed algorithms significantly outperform the intuitive algorithms that are based on spatial or temporal factors only.
Distributed virtual environment, time-space consistency, state update, scheduling.
Xueyan Tang, Suiping Zhou, "Update Scheduling for Improving Consistency in Distributed Virtual Environments", IEEE Transactions on Parallel & Distributed Systems, vol. 21, no. , pp. 765-777, June 2010, doi:10.1109/TPDS.2009.113
105 ms
(Ver 3.3 (11022016))