Distributed scalable encoder resources for live streams

Patent metadata

Field	Value
Publication number	US-10136165-B2
Application number	US-201113232870-A
Country	US
Kind code	B2
Filing date	Sep 14, 2011
Priority date	Sep 14, 2011
Publication date	Nov 20, 2018
Grant date	Nov 20, 2018

Abstract

Live media streams are delineated for distributed encoding and fragmentation in a dynamically scalable distributed resource system. In some examples, live MPEG-2 media streams are separated into groups of pictures (GOPs) and converted into jobs for encoding and fragmentation systems. Multiple jobs may be created for the same live media stream to provide for different quality levels for each fragment of each channel. Shared resources such as a pool of dynamically scalable virtual machines can be used to process the individual jobs. Encoded fragments may be H.264 fragments maintained on shared storage. A particular live stream corresponding to a particular quality level can be reconstructed using the encoded fragments.

First claim

The invention claimed is:

1. A method, comprising:
delineating, by a GOP delineator, a plurality of live streams including a first live stream into a plurality of groups of pictures (GOPs);
determining that a first group of pictures (GOP) is available from the GOP delineator;
placing a plurality of jobs associated with a first GOP into a plurality of job queues, the plurality of jobs corresponding to a plurality of encoding quality levels, the plurality of job queues accessible by a plurality of encoders and fragmentation systems implemented using a plurality of virtual machines in a cloud environment, wherein the cloud environment is dynamically scalable such that additional encoders and fragmentation systems are activated when resource usage in the cloud environment reaches a particular threshold and wherein the additional encoders and fragmentation systems are deactivated when resource usage falls below a particular threshold, wherein the plurality of job queues includes an active job queue and a standby job queue, each job queue being a first in first out queue;
performing encoding and fragmentation, prior to transmission to a user device, by the plurality of encoders and fragmentation systems at a first encoding quality level to generate a plurality of variants, including the first variant fragment and the second variant fragment;
transmitting, by the plurality of encoders and fragmentation system, the plurality of encoded fragments to distributed storage;
maintaining the plurality of encoded fragments in distributed storage, wherein a plurality of user devices playback the live media stream by continuously retrieving encoded fragments from distributed storage, wherein the plurality of encoded fragments are configured such that MPEG-4 files can be created without having to wait until all content is written to prepare movie headers, and wherein each fragment in the plurality of encoded fragments includes a box structure comprising synchronization information, end of file information, and chapter information in order to allow for MPEG-4 fragmentation without out of band signaling.

2. The method of claim 1 , wherein the GOP delineator resides in a first data center.

3. The method of claim 2 , wherein the plurality of encoders and fragmentation systems reside in a second data center.

4. The method of claim 3 , wherein the first GOP is encrypted prior to transmission to the second data center.

5. The method of claim 1 , wherein the plurality of encoders and fragmentation systems are dynamically scalable virtual machines.

6. The method of claim 5 , wherein an additional virtual machine is added to the plurality of virtual machines when processor resource usage exceeds a particular threshold.

7. The method of claim 5 , wherein an additional virtual machine is added to the plurality of virtual machines when the job queue level exceeds a particular threshold.

8. The method of claim 5 , wherein an additional virtual machine is added to the plurality of virtual machines when encoding delay exceeds a particular threshold.

9. The method of claim 1 , wherein the live media stream received at the GOP delineator is an MPEG-2 live media stream.

10. The method of claim 1 , wherein the plurality of encoded fragments is a plurality of MPEG-4 fragments.

11. A system, comprising:
a GOP delineator configured to delineate a plurality of live streams including a first live stream into a plurality of groups of pictures (GOPs);
a work dispatcher configured to determine that a first group of pictures (GOP) is available from the GOP delineator, wherein the work dispatcher is further configured to place a plurality of job associated with a plurality of encoding quality levels into plurality of job queues, wherein the plurality of job queues includes an active job queue and a standby job queue, each job queue being a first in first out queue;
a plurality of encoders and fragmentation systems implemented using a plurality of virtual machines in a cloud environment, wherein the cloud environment is dynamically scalable such that additional encoders and fragmentation systems are activated when resource usage in the cloud environment reaches a particular threshold and wherein the additional encoders and fragmentation systems are deactivated when resource usage falls below a particular threshold, the plurality of encoders and fragmentation systems operable to obtain the job from the plurality of job queues and perform encoding and fragmentation, prior to transmission to a user device, at a first encoding quality level to generate a plurality of variants, the plurality of encoders and fragmentation systems also operable to transmit the plurality of encoded fragments to distributed storage; a distributed storage system configured to maintain the plurality of encoded fragments, wherein a plurality of user devices playback the live media stream by continuously retrieving encoded fragments from distributed storage,
wherein the plurality of encoded fragments are configured such that MPEG-4 files can be created without having to wait until all content is written to prepare movie headers, and wherein each fragment in the plurality of encoded fragments includes a box structure comprising synchronization information, end of file information, and chapter information in order to allow for MPEG-4 fragmentation without out of band signaling.

12. The system of claim 11 , wherein the GOP delineator resides in a first data center.

13. The system of claim 12 , wherein the plurality of encoders and fragmentation systems reside in a second data center.

14. The system of claim 13 , wherein the first GOP is encrypted prior to transmission to the second data center.

15. The system of claim 11 , wherein the plurality of encoders and fragmentation systems are dynamically scalable virtual machines.

16. The system of claim 15 , wherein an additional virtual machine is added to the plurality of virtual machines when processor resource usage exceeds a particular threshold.

17. The system of claim 15 , wherein an additional virtual machine is added to the plurality of virtual machines when the job queue level exceeds a particular threshold.

18. The system of claim 15 , wherein an additional virtual machine is added to the plurality of virtual machines when encoding delay exceeds a particular threshold.

19. The system of claim 11 , wherein the live media stream received at the GOP delineator is an MPEG-2 live media stream.

20. A non-transitory computer readable medium comprising:
computer code for delineating, by a GOP delineator, a plurality of live streams including a first live stream into a plurality of groups of pictures (GOPs);
computer code for determining that a first group of pictures (GOP) is available from the GOP delineator;
computer code for placing a plurality of jobs associated with a first GOP into a plurality of job queues, the plurality of jobs corresponding to a plurality of encoding quality levels, the plurality of job queues accessible by a plurality of encoders and fragmentation systems implemented using a plurality of virtual machines in a cloud environment, wherein the cloud environment is dynamically scalable such that additional encoders and fragmentation systems are activated when resource usage in the cloud environment reaches a particular threshold and wherein the additional encoders and fragmentation systems are deactivated when resource usage falls below a particular threshold, wherein the plurality of job queues includes an active job queue and a

Assignees

• Karlsson Kent
• Barnes Fritz
• Wong Ian
• Haallmarker Ola
• Mobitv Inc

Inventors

• Karlsson Kent
• Barnes Fritz
• Wong Ian
• Hållmarker Ola

Classifications

• H04N21/2187

  Live feed · CPC title

• H04N21/8456

  by decomposing the content in the time domain, e.g. in time segments · CPC title

• H04N21/2181 Primary

  comprising remotely distributed storage units, e.g. when movies are replicated over a plurality of video servers (distributed storage of data in networks, e.g. transport arrangements for network file system [NFS], storage area networks [SAN] or network attached storage [NAS], H04L67/1097) · CPC title

• H04N21/23439

  for generating different versions · CPC title

• H04N21/6125

  involving transmission via Internet (transmission by internet of broadcast information H04H60/82) · CPC title