An Adaptive Load Balancing Technique for Distributed Virtual Environment Systems

One of the key issues in the design of scalable and costeffective Distributed Virtual Environment (DVE) systems is the partitioning problem. This problem consists of efficiently assigning clients (3-D avatars) to the servers in the system, and some methods have been already proposed for solving it. However, only one of these methods takes into account the non-linear behavior of DVE servers with the number of avatars they support, and this method uses a load balancing technique of local scope. As a result, it only provides good performance if the movement pattern of avatars is uniform. In this paper, we propose an adaptive load balancing technique of global scope for solving the partitioning problem in DVE systems. The global scope of the proposed technique allows to avoid DVE saturation as long as possible. Evaluation results show that the proposed strategy can improve DVE system performance, regardless of both the movement patterns of avatars and also the initial distribution of avatars in the virtual world.