Statistical admission control for soft real-time VOD servers

To fully utilize the resources with soft real-time video-ondemand servers, statistical admission control is widely used. There have been several models to capture the nature of bit rate variability of the video data. However, when the video server employs caching, the resulting workload exhibits different characteristics. In this paper, we present a probabilistic model to analyze the video server workload. Our model enables the quantitative analysis of the disk load reduction induced by caching so that the caching effect can be reflected jn the admission control. Using this model, it is possible to design a statistical admission control which provides a probabilistic guarantee for the jitter free playback.

[1]  Banu Özden,et al.  Buffer replacement algorithms for multimedia storage systems , 1996, Proceedings of the Third IEEE International Conference on Multimedia Computing and Systems.

[2]  Edward W. Knightly,et al.  Fundamental limits and tradeoffs of providing deterministic guarantees to VBR video traffic , 1995, SIGMETRICS '95/PERFORMANCE '95.

[3]  Heon Young Yeom,et al.  Exploiting caching for realtime multimedia systems , 1999, Proceedings IEEE International Conference on Multimedia Computing and Systems.

[4]  J. Walrand,et al.  RCBR: A Simple and Efficient Service for Multiple Time-Scale Traffic , 1995 .

[5]  Harrick M. Vin,et al.  A statistical admission control algorithm for multimedia servers , 1994, MULTIMEDIA '94.

[6]  Raymond Ng,et al.  Statistical Modeling and Buffer Allocation for Mpeg Streams , 1996 .

[7]  S. Keshav.  RCBR : A Simple and E cient Service for MultipleTime-Scale Tra , 1995 .

[8]  Banu Özden,et al.  Disk striping in video server environments , 1996, Proceedings of the Third IEEE International Conference on Multimedia Computing and Systems.