Methods and apparatuses for a distributed live-on-demand (LOD) origin

• US 9,819,972 B1
• Filed: 12/07/2015
• Issued: 11/14/2017
• Est. Priority Date: 12/10/2014
• Status: Active Grant

• Current Assignee: VideoLabs, Inc.
• Original Assignee: Digital Keystone Incorporated
• Inventors: Vantalon, Luc, Siccardo, Paolo
• Primary Examiner(s): Chokshi, Pinkal R
• Application Number:
  US14/961,848
• Time in Patent Office: 708 Days
• CPC Class Codes: H04N 21/23113 involving housekeeping operations for stored content, e.g. prioritizing content for deletion because of storage space restrictions storage management, e.g. defragmentation G06F3/0604; snloading stored programs G06F9/445; housekeeping operations in file systems, e.g. deletion policies G06F16/10; buffering arrangements in a network node or in an end terminal in packet networks H04L49/90
• CPC Class Codes: H04N 21/242 Synchronization processes, e.g. processing of PCR [Program Clock References] arrangements for synchronising broadcast or distribution via plural systems in broadcast distribution systems H04H20/18
• CPC Class Codes: H04N 21/262 Content or additional data distribution scheduling, e.g. sending additional data at off-peak times, updating software modules, calculating the carousel transmission frequency, delaying a video stream transmission, generating play-lists scheduling strategies for dispatcher in multiprogramming arrangements G06F9/4881; arrangements for scheduling broadcast services or broadcast-related services H04H60/06; flow control in packet networks H04L47/10; establishing a time schedule or organising the servicing of application requests H04L67/62
• CPC Class Codes: H04N 21/643 Communication protocols network streaming protocols, e.g. real-time transport protocol [RTP] or real-time control protocol [RTCP], H04L65/65
• CPC Class Codes: H04N 21/84 Generation or processing of descriptive data, e.g. content descriptors systems specially adapted for using meta-information in broadcast systems H04H60/73
• CPC Class Codes: H04N 21/8456 by decomposing the content in the time domain, e.g. in time segments
• CPC Class Codes: H04N 7/18 Closed-circuit television [CCTV] systems, i.e. systems in which the video signal is not broadcast

Abstract

A novel live on demand origin architecture in one embodiment can enable the provision, operation and management of massively scalable, highly efficient, high availability content access services that offer random access to linear feeds without playback restrictions, supporting live, near-live, catch-up, start-over and VOD access modes regardless of Origin, Network and Player limitations.

Claims

1. A method for continuously ingesting, recording and erasing multiple live video streams on a cluster of interconnected, autonomous live on demand nodes and delivering upon request any interval of the recordings, the method comprising:

ingesting the live video streams in a distributed manner on any one or more nodes of the cluster of interconnected, autonomous live on demand nodes from at least one source through a content acquisition network, the ingesting of the multiple live video streams being from multiple sources, each live video stream being ingested while the source of that live video stream is still receiving video content and wherein each live video stream is divided into multiple segment files, each segment file representing a unique format and bit rate of the live video stream for a fraction of the duration of a finite event in that live video stream, and each video stream has, while being received, an indefinite, unknown end time and an indefinite, unknown total length from start time to end time;

recording each of the live video streams on plurality of nodes of the cluster of interconnected, autonomous live on demand nodes, divided in multiple segment files, wherein each segment file is concurrently replicated on one or more live on demand nodes and is assigned a time-to-live value based on a per-stream recording profile and wherein each segment file is associated with multiple forms of metadata and wherein the segment files for each of the live video streams are recorded in a distributed manner across all of the nodes in the cluster of interconnected, autonomous live on demand nodes;

erasing, by each node, each instance of the recorded segment files on each node of the cluster of interconnected, autonomous live on demand nodes once they are older than their time-to-live value, wherein the erasing is scheduled to minimize disk and performance overheads based on the average segment file, wherein each segment file represents a unique format and bit rate of the live video stream for a fraction of a finite event in that live video stream;

upon an on demand request on any one or more nodes of the cluster of interconnected, autonomous live on demand nodes from one or more client devices, delivering through a content distribution network, to the one or more client devices either a) a list and description of all the segment files comprising a random interval of one of the recorded video streams, the list and description being constructed upon demand from the metadata of each segment file that composes the interval, wherein each segment file represents a unique format and bit rate of the live video stream for a fraction of a finite event in that live video stream or b) the segment files referenced in the transmitted list, wherein each segment file represents a unique format and bit rate of the live video stream for a fraction of a finite event in that live video stream, wherein the recorded video stream interval descriptions and segment files can be retrieved by the one or more client devices independently of when and where the recorded video stream segment files have been recorded, and wherein caching guidelines are dynamically created and appended to the response based on the time of the request, the time of the interval and the time-to-live of the segment file.

2. The method of claim 1 wherein the per-stream recording profile includes a stream identifier to be used for delivery requests, one or more source locators, a redundancy value, a redundancy mode and a recording time-to-live schedule, which defines based upon a set time period one or more time-to-live value transitions for the recording of that stream.

3. The method of claim 1 wherein the ingesting of each live video stream is performed by one or more participating nodes, the number of participating nodes being based on a redundancy value as defined in a per-stream recording profile.

4. The method of claim 3 wherein all the participating nodes are actively acquiring the segment files concurrently, and the method further comprising:

reading one or more forms of metadata to synchronize ingesting across all the participating nodes, wherein the metadata includes but is not limited to a calculated time reference and a measured duration for the last recorded segment files.

5. The method of claim 3 wherein one participating node is actively acquiring the segment files, while other participating nodes are monitoring the accumulation of one or more forms of metadata and wherein the method for the monitoring node further comprising:

detecting an interruption of the ingesting when no metadata has been recorded for longer than a predefined time period;

registering itself as the next active node after a random interval of time;

requesting the current active node to start monitoring, wherein the registering and the requesting are combined in a single atomic operation.

6. The method of claim 3 wherein the ingesting node records for each segment file multiple forms of metadata, prior to recording the file itself, the method further comprising:

recording with one or more searchable indexes, an absolute segment context, based on the segment description of the sourced live video stream packaging protocol, to support the transmitting upon request of a random interval starting from the segment in the same streaming protocol;

recording with one or more searchable indexes, one or more alternative segment contexts, based on the segment description of the sourced live video stream packaging protocol, to support the transmitting upon request of a random interval starting from the segment in one or more alternate streaming protocols;

recording with one or more searchable indexes multiple forms of performance details to support multi dimensional analytics queries.

7. The method of claim 1 wherein the erasing comprises calculating and providing compaction parameters to a compaction process, and wherein the calculation is a function of one or more average ingest rate, one or more average time-to-live values and the targeted disk overhead.

8. The method of claim 7 wherein the erasing is reconfigured when new recorded live video streams are added or removed, when recording time-to-live values change in each stream schedule, or when changes in the Content Acquisition Network modify the structure of each ingested live video stream.

9. The method of claim 7 wherein the erasing is sequenced by grouping all the live stream recordings into one or more groups, and by providing each group with a different variation of at least one of the nominal parameter values.

10. The method of claim 7 wherein the erasing is further optimized by calculating and providing different parameter values per categories of video streams that share similar recording profiles, wherein the calculation is using a different average ingest rate and a different average time-to-live value per category.

11. The method of claim 1 wherein the on demand request can occur at any time after the start of the recording of the live video streams but prior to erasing the live video stream segment files.

12. The method of claim 1 wherein the list and description of the segment files for a random interval of one of the live video streams depends on the requested streaming protocol, the time of ingesting, the start and duration of the interval, the time of the request, a detection of missing segment files and a presence of singular segments.

13. The method of claim 11 wherein a segment is singular because it has been associated in the sourced live video streaming protocol with an in-band event, including but not limited to events identified according to a SCTE 35 protocol.

14. The method of claim 11 wherein the caching guidelines for downstream content delivery networks are calculated based on one or more of the requested streaming protocol, the start and duration of the interval, the time of the request, the segment duration, the remaining time-to-live of the segment files and the overall response time.

15. The method of claim 1 wherein the segments and their multiple forms of metadata are recorded in one or more tables of a distributed non-relational database.

16. The method of claim 1 wherein the ingesting is performed on one or more ingesting nodes different than the live on demand nodes.

17. The method of claim 1 wherein the on-demand request delivering is performed on one or more delivery nodes different than the live on demand nodes.

18. A cluster of interconnected data processing systems, which act as nodes to perform the method of claim 1.

19. The method of claim 1 wherein one of the multiple sources is one of the LOD nodes, wherein the live video stream corresponds to a previously recorded interval, and wherein the per stream recorded profile creates a new copy of the interval with extended time-to-live value.

20. One or more non-transitory machine readable media storing executable program instructions that, when executed by one or more data processing systems, cause the one or more data processing systems to perform a method for continuously ingesting, recording and erasing multiple live video streams on a cluster of interconnected, autonomous live on demand nodes and delivering upon request any interval of the recordings, the method comprising:

ingesting the live video streams in a distributed manner on any one or more nodes of the cluster of interconnected, autonomous live on demand nodes from at least one source through a content acquisition network, the ingesting of the multiple live video streams being from multiple sources, each live video stream being ingested while the source of that live video stream is still receiving video content and wherein each live video stream is divided into multiple segment files, each segment file representing a unique format and bit rate of the live video stream for a fraction of the duration of a finite event in that live video stream, and each video stream has, while being received, an indefinite, unknown end time and an indefinite, unknown total length from start time to end time;

recording each of the live video streams on plurality of nodes of the cluster of interconnected, autonomous live on demand nodes, divided in multiple segment files, wherein each segment file is concurrently replicated on one or more live on demand nodes and is assigned a time-to-live value based on a per-stream recording profile and wherein each segment file is associated with multiple forms of metadata and wherein the segment files for each of the live video streams are recorded in a distributed manner across all of the nodes in the cluster of interconnected, autonomous live on demand nodes;

erasing, by each node, each instance of the recorded segment files on each node of the cluster of interconnected, autonomous live on demand nodes once they are older than their time-to-live value, wherein the erasing is scheduled to minimize disk and performance overheads based on the average segment file, wherein each segment file represents a unique format and bit rate of the live video stream for a fraction of a finite event in that live video stream;

upon an on demand request on any one or more nodes of the cluster of interconnected, autonomous live on demand nodes from one or more client devices, delivering through a content distribution network, to the one or more client devices either a) a list and description of all the segment files comprising a random interval of one of the recorded video streams, the list and description being constructed upon demand from the metadata of each segment file that composes the interval, wherein each segment file represents a unique format and bit rate of the live video stream for a fraction of a finite event in that live video stream or b) the segment files referenced in the transmitted list, wherein each segment file represents a unique format and bit rate of the live video stream for a fraction of a finite event in that live video stream, wherein the recorded video stream interval descriptions and segment files can be retrieved by the one or more client devices independently of when and where the recorded video stream segment files have been recorded, and wherein caching guidelines are dynamically created and appended to the response based on the time of the request, the time of the interval and the time-to-live of the segment file.

21. The one or more media of claim 20 wherein the per-stream recording profile includes a stream identifier to be used for delivery requests, one or more source locators, a redundancy value, a redundancy mode and a recording time-to-live schedule, which defines based upon a set time period one or more time-to-live value transitions for the recording of that stream.

22. The one or more media of claim 20 wherein the ingesting of each live video stream is performed by one or more participating nodes, the number of participating nodes being based on a redundancy value as defined in a per-stream recording profile.

23. The one or more media of claim 22 wherein all the participating nodes are actively acquiring the segment files concurrently, and the method further comprising:

reading one or more forms of metadata to synchronize ingesting across all the participating nodes, wherein the metadata includes but is not limited to a calculated time reference and a measured duration for the last recorded segment files.

24. The one or more media of claim 22 wherein one participating node is actively acquiring the segment files, while other participating nodes are monitoring the accumulation of one or more forms of metadata and wherein the method for the monitoring node further comprising:

detecting an interruption of the ingesting when no metadata has been recorded for longer than a predefined time period;

registering itself as the next active node after a random interval of time;

requesting the current active node to start monitoring, wherein the registering and the requesting are combined in a single atomic operation.

25. The one or more media of claim 22 wherein the ingesting node records for each segment file multiple forms of metadata, prior to recording the file itself, the method further comprising:

recording with one or more searchable indexes, an absolute segment context, based on the segment description of the sourced live video stream packaging protocol, to support the transmitting upon request of a random interval starting from the segment in the same streaming protocol;

recording with one or more searchable indexes, one or more alternative segment contexts, based on the segment description of the sourced live video stream packaging protocol, to support the transmitting upon request of a random interval starting from the segment in one or more alternate streaming protocols;

recording with one or more searchable indexes multiple forms of performance details to support multi-dimensional analytics queries.

26. The one or more media of claim 20 wherein the erasing is a two-step process, the method further comprising:

obsoleting segment files immediately upon expiration across all nodes;

configuring each node dynamically to progressively and independently erase all the obsolete segment files based on current ingesting activity and the recording schedule of each live video streams.

27. The one or more media of claim 26 wherein the erasing is optimized per group of live video stream sharing the same recording schedule.

28. The one or more media of claim 20 wherein the on demand request can occur at any time after the start of the recording of the live video streams but prior to erasing the live video stream segment files.

29. The one or more media of claim 20 wherein the list and description of the segment files for a random interval of one of the live video streams depends on the requested streaming protocol, the time of ingesting, the start and duration of the interval, the time of the request, a detection of missing segment files and a presence of singular segments.

30. The one or more media of claim 28 wherein a segment is singular because it has been associated in the sourced live video streaming protocol with an in-band event, including but not limited to events identified according to a SCTE 35 protocol.

31. The one or more media of claim 28 wherein the caching guidelines for downstream content delivery networks are calculated based on the requested streaming protocol, the start and duration of the interval, the time of the request, the remaining time-to-live of the segment files and the overall response time.

32. The one or more media of claim 20 wherein the segments and their multiple forms of metadata are recorded in one or more tables of a distributed non-relational database.

33. A cluster of autonomous live on demand nodes, each of which has a non-transitory machine readable storage medium that is one of the one or more media of claim 20.