Distributed Computing for Real-time Video Processing

Real-time video processing, such as rendering and/or encoding, is one of the most bandwidth demanding operation in audio-visual applications [1][2]. There have been many efforts at parallelizing video rendering and encoding architecture mainly for VLSI chip design [3][4][5][6][7][8][9][10][11][12][13] [14][15][16]. However, not a lot of research has been dedicated in the area of distributed computing for real-time video processing [1][2][17][18][19][20]. Distributed computing on the Internet has mainly been applied to applications in which data could be processed in non-real-time [10]. For visual communication, real-time constraints give additional requirements to data processing in distributed computing. It is necessary to assure the processing time of distributed data since processing period for one frame of video is limited to 1/25 or 1/30 second in most cases. Thus, processing delay is a critical factor for video processing applications especially in the case of non-homogeneous computing environment, such as distributed computing on the Internet. In this paper, we first survey the trend of distributed computing for video processing, followed by introducing some architecture for distributed computing of real-time video application, particularly video encoding.