1. Scalable On-demand Media Streaming Anirban Mahanti mahanti@cpsc.ucalgary.ca Department of Computer Science University of Calgary Canada T2N 1N4

2. 2 Introduction Context: Video-on-demand applications on the Internet, satellite & cable television networks E.g., Online courses, movies, interactive TV Goals: Scalable delivery Sub-linear server/network bandwidth

3. 3 Video-on-Demand Distribution Model A client can tune in to receive any ongoing media delivery using its Set Top Box True broadcast: Satellite and cable TV networks Multipoint delivery provided in the Internet by IP- Multicast or Application Layer Multicast

4. 4 Traffic Assumptions 100s – 1000s requests for a media file per play duration Skewed popularity of media files 10% – 20% of the files account for 80% of the requests

5. 5 Download vs. Streaming Download: Receive entire content before playback begins High “start-up” delay as media file can be large ~ 4GB for a 2 hour MPEG II movie Streaming: Play the media file while it is being received Reasonable “start-up” delays True video-on-demand service Reception Rate >= playback rate

6. 6 Scalable Streaming: Motivation Consider a popular media file Playback rate: 1 Mbps Duration: 90 minutes Request rate: once every minute Start a new stream at the playback rate for each request: Bandwidth required = 1 Mbps x 90 Leverage the multipoint delivery capability of modern networks

7. 7 Scalable Streaming Protocols: Overview Bounded Delay Protocols Batching, Periodic Broadcasts Tradeoff: start-up delay vs. bandwidth Immediate Service Protocols Patching, Bandwidth Skimming Tradeoff: request rate vs. bandwidth

8. 8 Batching Example Playback rate = 1 Mbps, duration = 90 minutes Group requests in non-overlapping intervals of 30 minutes: Max. start-up delay = 30 minutes Bandwidth required = 3 channels = 3 Mbps Bandwidth increases linearly with decrease in start-up delay 03030 6090120150180210240 Time (minutes) Channel 1 Channel 2 Channel 3

9. 9 Periodic Broadcast Example Partition the media file into 2 segments with relative sizes {1, 2}. For a 90 min. movie: Segment 1 = 30 minutes, Segment 2 = 60 minutes Advantage: Max. start-up delay = 30 minutes Bandwidth required = 2 channels = 2 Mbps Disadvantage: Requires increased client capabilities Time (minutes) 1 2 1 1 11 1 22 0306090120150180 Channel 1 Channel 2

10. 10 Skyscraper Broadcasts (SB) Divide the file into K segments of increasing size Segment size progression: 1, 2, 2, 5, 5, 12, 12, 25, … Multicast each segment on a separate channel at the playback rate Aggregate rate to clients: 2 x playback rate [Hua & Sheu 1997]

11. 11 Comparing Batching and SB Server Bandwidth Start-up Delay BatchingSB 1 Mbps90 minutes 2 Mbps45 minutes30 minutes 6 Mbps15 minutes3 minutes 10 Mbps9 minutes30 seconds Playback rate = 1 Mbps, duration = 90 minutes Limitations of Skyscraper: Ad hoc segment size progress Does not work for low client data rates

12. 12 Reliable Periodic Broadcasts (RPB) Optimized PB protocols (no packet loss recovery) client fully downloads each segment before playing required server bandwidth near minimal Segment size progression is not ad hoc Works for client data rates < 2 x playback rate extend for packet loss recovery extend for “bursty” packet loss extend for client heterogeneity We will not cover this part [Mahanti et al. 2001, 2003, 2004]

13. 13 Optimized Periodic Broadcasts r = segment streaming rate = 1 s = maximum # streams client listens to concurrently = 2 b = client data rate = s x r = 2 length of first s segments: length of segment k  s:

14. 14 Optimized PB: Performance r = segment transmission rate, s = max. # streams client listens to concurrently b = client data rate = s x r

15. 15 Scalable Streaming Protocols … Bounded Delay Protocols Batching, Periodic Broadcasts Tradeoff: start-up delay vs. bandwidth Immediate Service Protocols Patching, Bandwidth Skimming Tradeoff: request rate vs. bandwidth

16. 16 Patching Clients use a “patch” stream to catch-up with the “root” stream Server Bandwidth scales as square root [Carter & Long 1997, Hua et al. 1998]

17. 17 Bandwidth Skimming Allocate a multicast stream to each client; a client also listens to closest earliest active stream Bandwidth scales logarithmically [Eager et al. 1999]

18. 18 Bandwidth Skimming: Performance Bandwidth Skimming better than Patching Bandwidth Skimming policies allow merging for b < 2

19. 19 Summary Discussed some techniques for scalable on- demand media streaming Bounded delay protocols Immediate service protocols Reliable Delivery? Non-linear media? Non-sequential media access?

20. 20 For Details … Anirban Mahanti, Scalable Reliable On-Demand Media Streaming Protocols, Ph.D. Thesis, Dept. of Computer Science, Univ. of Saskatchewan, March 2004. Anirban Mahanti, Derek L. Eager, Mary K. Vernon, David Sundaram-Stukel, Scalable On-Demand Media Streaming with Packet Loss Recovery, IEEE/ACM Trans. On Networking, April 2003. Also in ACM SIGCOMM 2001. Email: mahanti@cs.ucalgary.camahanti@cs.ucalgary.ca http://pages.cpsc.ucalgary.ca/~mahanti