The Scalable Adaptive Graphics Environment (SAGE) is a specialized middleware that enables real-time streaming of extremely high-resolution graphics and high-definition video from remotely distributed rendering and storage clusters to scalable display walls over ultra high-speed networks. I propose Visualcasting in order to extend SAGE to support distant collaboration between multiple endpoints by streaming visualization to all participants.

In the SAGE framework, each visualization application streams its rendered pixels to the virtual high-resolution frame buffer of SAGE, allowing users to freely move, resize and overlap the application windows on the display. Every window movement or resize operation requires dynamic and non-trivial reconfigurations of the involved graphics streams. These reconfigurations become even more complex when SAGE is required to support multiple collaboration endpoints with different tiled display configurations and application window layouts.

Visualcasting is a scalable real-time image distribution service to address this problem. Visualcasting includes a high-speed bridging system that receives pixel streams from rendering clusters to duplicate and distribute pixels to each end-point. This approach will be compared with the traditional router-based multicasting approach and the reliable layered multicasting approach. Several issues in optimizing Visualcasting will be also addressed.

PDF

Jeong, B., Visualcasting - Scalable Real-Time Image Distribution in Ultra-High Resolution Display Environments, Electronic Visualization Lab - Preliminary Prospectus, July 15th, 2006.