1. Video Staging: A Proxy-Server-Based Approach to End-to-End Video Delivery over Wide-Area Networks Zhi-Li Zhang, Yuewei Wang, David H. C. Du, Dongli Su IEEE/ACM TRANSACTIONS ON NETWORKING, AUGUEST 2000

2. Outline n Introduction n Video Staging: A Single Video Case –without Smoothing –with Smoothing –Empirical Evaluation n Video Staging: Multiple Video Case –Staging Hot Video Only(SHVO) –Largest Bandwidth Reduction Ratio First(LBRRF) –Empirical Evaluation n Conclusion

3. Introduction n Video Staging –Reduce the bandwidth requirement in the Backbone WAN –Prefetch a predetermined amount of video data and store them a priori at proxy servers –Only part of a video stream is retrieved directly from the central video server across the backbone WAN

4. Video Staging(cont.) –Trading the disk bandwidth of a proxy server for the backbone WAN bandwidth –The video data are staged at the proxy server on a fairly long period of time instead of caching in and purged out dynamically

6. Video Staging:A Single Video Case n Video Staging Without Smoothing i : the index of a video N i : total number of frames F : frame period s i j : the size of jth frame j = 1,…, N i P i : the peak rate of video i P i = (max 1<=j<=Ni s i j ) /F C i : cut-off rate 0 <= C i <= P i *F = max 1<=j<=Ni s i j

8. Video Staging Without Smoothing n The upper part consists of a sequence of partial frames with size s i j,u = ( s i j - C i ) + j = 1,…, N i where x + = max{x,0} n The lower part consists of a sequence of partial frames with size s i j,l = s i j - ( s i j - C i ) + j = 1,…, N i

9. Video Staging Without Smoothing n The smaller C i is,the more video data is staged at a proxy server. n As C i decreases, the lower part of the video becomes less bursty, and eventually approaches to an essentially CBR stream. n The backbone WAN bandwidth requirement from P i to T i = C i / F n The upper part of the video consume D i = (max 1<=j<=Ni s i j,u ) / F disk bandwidth n Bandwidth reduction ratio R i = (P i - T i ) / D i

10. Video Staging With Smoothing n Cut-Off After Smoothing (CAS) s i ~j : referred to as smoothed frames size P i : the peak rate of smoothed stream C i : cut-off rate 0 <= C i <= P i *F = max 1<=j<=Ni s i ~j T i = C i / F amount of backbone WAN bandwidth is reserved D i = (max 1<=j<=Ni s i ~j,u ) / F amount of disk bandwidth is required in the worst case

11. Video Staging With Smoothing n Cut-Off Before Smoothing –Smoothing on the Lower Part(SOLP) T i = P i = (max 1<=j<=Ni s i ~j,l ) / F –Smoothing on the Upper Part(SOUP) Reduce the disk bandwidth required –Smoothing on the Upper and Lower Parts(SOLP) B l = B * (C i / P i * F) B u = B * (1- (C i / P i * F))

12. Proxy server disk resource requirements for a single stream

13. Resource requirements for a single stream

14. Ratio of backbone WAN bandwidth reduction to proxy server disk bandwidth

15. Video Staging: Multiple Video Case n Disk bandwidth n Total amount of storage space n The number of expected concurrent accesses n Total reduction in the backbone WAN bandwidth is maximized n Disk bandwidth constraint n Disk storage constraint

16. Staging Hot Video Only (SHVO) n Regard the user access pattern as the most important factor in determining which video to stage at the proxy server. n The videos are ordered with respect to their relative “popularity” n The number of k hottest videos that can be staged at the proxy server is determined n Either a video is entirely staged at the proxy server or not at all

17. Largest Bandwidth Reduction Ratio First (LBRRF) n Use the backbone WAN bandwidth reduction ratio R i in determining which video and what percentage of it to be staged n The burstier video and the video with larger number of concurrent accesses is more likely to have a higher R i n CP i percentage of video i is staged at the proxy server n The algorithm starts with CP i = 0 for all videos and iterates by incrementing CP i by  P amount at each step.

19. Impact of proxy server disk resources: clients have no smoothing buffer

20. Impact of proxy server disk resources: clients have 64KB smoothing buffer

21. Conclusion n A major objective of our approach is to reduce the backbone WAN bandwidth requirement. n We have designed various video staging methods and evaluated their effectiveness in trading the disk bandwidth of a proxy server for the backbone WAN bandwidth. n The proposed proxy-server-based video staging technique provides a cost-effective and scalable solution to the problem of the end-to-end video delivery over WANs.