Return to search

An Architecture Design for a Real-Time Web-Based Visualization in the Grid Environment

Situations like war, terrorist attacks, fire accidents, floods, storms, etc., which threaten human life and property, demand immediate action to decrease the damage caused by them. A system is needed that predicts the future events based on what has happened and notifies the concerned personnel. The situation could be better understood in less time if the data is represented as colored, shaded and moving images rather than as numbers. Such a system requires a real-time Web-based visualization system with easy and secure access to grid resources, presenting easy-to-read graphics through a simple interface provided by a Web browser, and responding to user actions immediately. The Web and grid environments impose severe performance constraints such as communication time, latency of the network, etc., making it highly difficult to have a highly responsive real-time visualization. This work aims in finding an appropriate design that satisfies the above requirements. It also aims in understanding the limitations of a distributed environment for real-time applications and finding ways to overcome those limitations. A three-tier architecture is proposed, implemented, and tested to find the bottlenecks of the distributed environment. Relevant design principles are applied to a case study eliminating or minimizing the bottlenecks until the case study system satisfies all the requirements. The case study is the Fire-Smoke system, simulating the propagation of fire in the ex-USS Shadwell test area emulating a submarine. This system is re-implemented from a stand-alone system in OpenGL to a real-time Web-based visualization system using Java3D and J2EE technologies.

Identiferoai:union.ndltd.org:MSSTATE/oai:scholarsjunction.msstate.edu:td-1480
Date11 December 2004
CreatorsSura, Bhargavi
PublisherScholars Junction
Source SetsMississippi State University
Detected LanguageEnglish
Typetext
Formatapplication/pdf
SourceTheses and Dissertations

Page generated in 0.0021 seconds