MEDIA STREAMING WITH ONLINE CACHING AND PEER-TO-PEER FORWARDING

• US 20090055471A1
• Filed: 08/19/2008
• Published: 02/26/2009
• Est. Priority Date: 08/21/2007
• Status: Active Grant

• Current Assignee: NTT Docomo Incorporated (Nippon Telegraph and Telephone Corporation)
• Original Assignee: NTT Docomo Incorporated (Nippon Telegraph and Telephone Corporation)
• Inventors: Harmanci, Oztan, Demircin, Mehmet U., Kozat, Ulas C., Kanumuri, Sandeep
• Granted Patent: US 8,078,729 B2
• Time in Patent Office: Days
• US Class Current: 709/203
• CPC Class Codes: H04L 65/612 for unicast
• CPC Class Codes: H04L 67/104 Peer-to-peer [P2P] networks
• CPC Class Codes: H04L 67/1063 Discovery through centralising entities
• CPC Class Codes: H04L 67/1076 Resource dissemination mechanisms or network resource keeping policies for optimal resource availability in the overlay network
• CPC Class Codes: H04L 67/108 characterised by resources being split in blocks or fragments
• CPC Class Codes: H04L 67/1091 Interfacing with client-server systems or between P2P systems
• CPC Class Codes: H04L 67/56 Provisioning of proxy services store-and-forward switching systems in data switching networks H04L12/54
• CPC Class Codes: H04L 67/5682 Policies or rules for updating, deleting or replacing the stored data
• CPC Class Codes: Y10S 707/967 Peer-to-peer

Abstract

A system, method and apparatus are disclosed herein for media streaming. In one embodiment, the system comprises one or more media servers to serve media content and a plurality of peers communicably coupled to one or more other peers of the plurality of peers and at least one of the one or more media servers to receive segments of media content, where at least one of peers allocates a set of resources for serving the segments of media content including cache memory to store the segments and media files and uplink bandwidth to send the segments of media content to the one or more peers to which the one peer is communicably coupled. The system also includes a first control server to track media content demand and the allocated resources of the plurality of peers to determine which peer should cache which segment of the media file and to return peer location information specifying the one or more peer locations from which each peer is to receive each segment of each media content requested. The control server is operable to send the location information to each peer. In one embodiment, the one control server is also operable to calculate a utility of each caching option and enforce it by sending triggers to the peers to initiate the caching at those peers.

Claims

1. A system comprising:

one or more media servers to serve media content;

a plurality of peers, communicably coupled to one or more other peers of the plurality of peers and at least one of the one or more media servers, to receive segments of media content, wherein at least one of the plurality of peers allocates a set of resources for serving the segments of media content including cache memory to store the segments and media files and uplink bandwidth to send the segments of media content to the one or more peers to which the one peer is communicably coupled; and

a first control server to track media content demand and the allocated resources of the plurality of peers to determine peer location information specifying the one or more locations from which each peer is to receive each segment of each media content requested, the at least one control server operable to send the location information to said each peer.

2. The system defined in claim 1 wherein the first control server tracks supply and demand of each segment of media content by each peer.

3. The system defined in claim 1 wherein the first control server determines which peer location should cache which video segment using peer arrival rates and supply-demand analysis with respect to the peers and supply and demand of segments of media content with respect to the peers.

4. The system defined in claim 1 wherein the first control server determines the peer location information by attempting to maximize uplink bandwidth utilization at each peer by making desired segments of media content accessible at a particular time.

5. The system defined in claim 1 wherein the media is partitioned into contiguous segments and the first control server computes an estimated future demand for each of the segments.

6. The system defined in claim 1 wherein the first control server causes a segment of media to be streamed to one of the plurality of peers after the one peer joins the system, the one peer operable to cache the segment into its cache memory.

7. The system defined in claim 6 wherein one of both the first control server and the one peer determine whether the one peer continues to store the segment or overwrites the segment with a new segment being downloaded for playback.

8. The system defined in claim 1 wherein determining whether to cache the new segment is based on a determined amount of reduction in load of at least one of the one or more media servers achieved over a period of time.

9. The system defined in claim 1 wherein determining whether to cache the new segment is based on a prediction of future demand of the new segment and capability of peers and media servers to supply the new segment to other peers.

10. The system defined in claim 1 wherein the first control server controls access to one video during time the video is available for playback in the system.

11. The system defined in claim 1 wherein each peer has a cache memory limited in size to storing a set number of segments of media.

12. The system defined in claim 1 wherein the first control server is operable to estimate supply and demand curves corresponding to each segment of media content at a future time and use each estimate to determine the location information.

13. The system defined in claim 12 wherein the first control server generates the estimate using peer arrival and departure time statistics.

14. The system defined in claim 12 wherein the first control server generates the estimate using peer arrival and departure time statistics, information indicative of when a particular media segment is requested, node inter-arrival and inter-departure statistics, information about round-trip-time communication and processing delays.

15. The system defined in claim 12 wherein the first control server generates the estimate using one or more of a group consisting of media playback rate, media download rates, and media segment sizes.

16. The system defined in claim 12 wherein the estimates are updated in regular or irregular intervals.

17. The system defined in claim 12 wherein the first control server generates the estimate based on a utility measure computed for each media file and each segment of a media file from supply and demand curves corresponding to the supply and demand with respect to said each segment.

18. The system defined in claim 17 wherein the first control server determines which segments to cache based on computed utility functions associated with each segment.

19. The system defined in claim 17 wherein the utility function is a function applied to a difference between an estimated total demanded bit rate for said each segment for a future time instance estimated at a current time instance and an estimated total upstream bandwidth of peers caching said each segment for the future time instance estimated at the current time instance.

20. The system defined in claim 19 wherein the function comprises one selected from a group consisting of:

$I<br><br>\left[x\right]=\{\begin{array}{c}x,x><br><br>0\\ 0<br><br><br><br>x\le <br><br>0\end{array};<br><br><br><br>I<br><br>\left[x\right]=\{\begin{array}{c}1,x><br><br>0\\ 0<br><br><br><br>x\le <br><br>0\end{array};<br><br><br><br>I<br><br>\left[x\right]=x;<br><br>\mathrm{and}<br><br><br><br>I<br><br>\left[x\right]=\{\begin{array}{cc}0,& x\le <br><br>0\\ x,& R_u^y\ge <br><br>x><br><br>0\\ R_u^y,& x><br><br>R_u^y\end{array},$

where R_u^y is the upload rate of user y; and

wherein x represents the difference.

21. The system defined in claim 1 wherein the first control server is operable to send triggers to a peer to start caching process for one or more segments of media content.

22. The system defined in claim 1 wherein the first control server is operable to determine how media files are segmented and determine download rates and playback delays of each segment.

23. The system defined in claim 1 wherein at least one peer specifies locally available resources to the first control server.

24. The system defined in claim 1 wherein one of the plurality of peers is operable to receive a trigger from the first control server to cache one or more segments of media content.

25. The system defined in claim 24 wherein the one peer pre-fetches the one or more segments in response to the trigger from the one or more media servers and one or more other peers.

26. The system defined in claim 24 wherein the one peer receives the one or more segments to be played immediately or in the future.

27. The system defined in claim 1 where the control server sends one or more control messages to at least one of a group of a media server and a peer, to cause the at least one media server or peer to push a video segment to a peer requesting the video segment.

34. The system defined in claim 8 further comprising determining how media files are segmented and determine download rates and playback delays of each segment.

28. A method comprising:

tracking media content demand and allocated resources of a plurality of peers to determine location information specifying the one or more locations from which each peer is to receive each segment of each media content requested; and

sending the location information to said each peer.

29. The method defined in claim 28 further comprising tracking supply and demand of each segment of media content by each peer.

30. The method defined in claim 28 further comprising determining the which peer location to cache which video segment using peer arrival rates and supply-demand analysis with respect to the peers and supply and demand of segments of media content with respect to the peers.

31. The method defined in claim 28 further comprising determining the sending peer location information by attempting to maximize uplink bandwidth utilization at each peer by making desired segments of media content accessible at a particular time.

32. The method defined in claim 28 further comprising determining whether to cache a new segment based on one or more of a group consisting of:

a determined amount of reduction in load of at least one media server achieved over a period of time and a prediction of future demand of the new segment, and capability of peers and media servers to supply the new segment to other peers.

33. The method defined in claim 28 further comprising estimating supply and demand curves corresponding to each segment of media content at a future time and using each estimate to determine the location information, wherein estimating supply and demand curves is performed using one or more of a group consisting of:

peer arrival and departure time statistics, information indicative of when a particular media segment is requested, node inter-arrival and inter-departure statistics, information about round-trip-time communication and processing delays, media playback rate, media download rates, and media segment sizes, a utility measure computed for each media file and each segment of a media file from supply and demand curves corresponding to the supply and demand with respect to said each segment.