LOADER: A Location-Aware Distributed Virtual Environment Architecture

Abstract

This article proposes a new architecture for distributed data management and update massage exchange in massively multi-user virtual environments. The key points in the design of such environments are scalability and QoS-aware message delivery. Therefore it requires robust distributed algorithms in a dynamic peer-to-peer system with frequent node arrivals and departures. Our proposed approach is mainly based on distributed hash tables (DHTs) in order to achieve a decentralized system where any participating node can efficiently retrieve the updates associated with a given object. Therefore, the responsibility of maintaining the mapping from names to values is distributed among the nodes in a way that a change in the set of participants causes a minimal amount of disruption. This allows the design to scale to extremely large number of nodes and to handle continual node arrivals, departures, and failures. The original form of DHT has several deficiencies for multi-user virtual environment application. This work introduces LOADER: a Location Aware Distributed Virtual Environment Architecture, which is the result of a series of efforts in alleviating a number of these problems including lack of location-aware ID assignment. Moreover, efficient multicast is addressed as an inevitable requirement for message exchange in multi-user environments and a solution based on Scribe is proposed.