This paper deals with the design and implementation of a low cost, distributed Virtual Reality system. The conceptual model of the system addresses the dependency of the visual update rate on simulation computation. The Process Agent Model is introduced to make the details of distributed processing and network communications transparent to the high level processes. Processes residing on the same machine communicate with one another directly by means of method calls. `Agents' are used for communication between remote processes. An agent intercepts messages on behalf of the process which it represents, and via the network these messages are forwarded to the authentic remote process. A network protocol which is suited to real-time VR applications is discussed, allowing for the transmission of both guaranteed and non-guaranteed message types. The network employs an intelligent queuing mechanism, based on message type and target process, for the filtering of redundant and aged messages. This filtering contributes to a reduction of bandwidth utilization. A qualitative performance analysis of the prototype system reveals three factors influencing system performance: division of computational load, network lag and system overhead.
CITATION STYLE
Blumenow, W., Spanellis, G., & Dwolatzky, B. (1997). Process agent model and message passing in a distributed processing VR system. In ACM Symposium on Virtual Reality Software and Technology, Proceedings, VRST (pp. 165–172). ACM. https://doi.org/10.1145/261135.261166
Mendeley helps you to discover research relevant for your work.