Using Multi-threading and Server Update Pushing to Improve the Performance of VNC for a Wall-Sized Tiled Display Wall

Display walls are wall-sized, high-resolution displays, typically built using several computers, each driving a projector or an LCD. The VNC (Virtual Network Computer) model is a simple way of creating desktops large enough for display walls by using a centralized virtual frame buffer. However, performance suffers significantly when the resolution increases due to the centralized server locating and compressing updates for the display computers. Another problem is that the display computers request and receive updates independently, resulting in an inconsistent view. TiledVNC is developed to better adapt VNC to a display wall and improve performance over an existing implementation, TightVNC. The changes include multi-threading, a server push update protocol, and pushing updates for the same frame to all viewers. To evaluate our system, we play two videos on our 22 megapixel display wall. Compared to TightVNC, TiledVNC increases the frame rate with up to 46% for a 6.75 megapixel video.

[1]  Anoop Gupta,et al.  Tools and applications for large-scale display walls , 2005, IEEE Computer Graphics and Applications.

[2]  Yong Liu,et al.  MultiStream A Cross-Platform Display Sharing System Using Multiple Video Streams , 2008, 2008 The 28th International Conference on Distributed Computing Systems Workshops.

[3]  Yong Liu,et al.  Improving the performance of VNC for high-resolution display walls , 2009, 2009 International Symposium on Collaborative Technologies and Systems.

[4]  Greg Humphreys,et al.  Chromium: a stream-processing framework for interactive rendering on clusters , 2002, SIGGRAPH.

[5]  David R. Butenhof Programming with POSIX threads , 1993 .

[6]  Andy Hopper,et al.  Virtual Network Computing , 1998, IEEE Internet Comput..

[7]  Andrew E. Johnson,et al.  High-Performance Dynamic Graphics Streaming for Scalable Adaptive Graphics Environment , 2006, ACM/IEEE SC 2006 Conference (SC'06).

[8]  Jason Nieh,et al.  Measuring thin-client performance using slow-motion benchmarking , 2001, TOCS.

[9]  John Markus Bjørndalen,et al.  Gesture-Based, Touch-Free Multi-User Gaming on Wall-Sized, High-Resolution Tiled Displays , 2008, J. Virtual Real. Broadcast..