MULTI-SCALE DATA SETS IN DISTRIBUTED ENVIRONMENTS

Within the scope of the research project "NEXUS" that is being worked on at the University of Stuttgart in cooperation between the Institute for Photogrammetry, the Institute of Parallel and Distributed High-Performance Systems and the Insitute of Communication Networks and Computer Engineering, a generic platform supporting location aware applica- tions with mobile users will be developed. The central component of the infrastructure consists of dynamic spatial mod- els. Since huge amounts of data have to be handled by the platform, they have to be distributed on multiple servers. Furthermore, the spatial data have to be delivered in multiple representations in order to provide the adequate level of detail for each application. The spatially aware applications which are using the spatial models are available for the users on small handheld com- puters. These computers are amongst others equipped with sensors for positioning and components for mobile commu- nication. Therefore, the applications know the position of the users within the spatial models - they are "location aware". Moreover, the users can receive spatial data from distributed model servers by means of wireless communica- tion. Thus, they can not only move through the real but also through the virtual world. In order to support the users on their way, GIS functionality has to be provided: they have to be able to perform spatial queries on their environment and to navigate themselves through the virtual models and the real world, respectively. The paper will give an introduction to the general concepts of the project. It presents the approaches concerning multi- ple representations and describes the strategies for distributed data handling that will be applied within the platform, including some remarks on prototypical solutions.