Granted Patent B1

US 12,192,496 · App. 17/956,483 · Granted Jan 7, 2025

Inventors: Ryan Hegar (Happy Valley, OR); Eric Woodruff (Portland, OR); David William Cardwell (Portland, OR); Brian Enigma (Portland, OR); Karthik Bangalore Shamsunder (Portland, OR); Darin J. Klaas (Beaverton, OR)

Assignee: Amazon Technologies, Inc.

H04N19/40H04N21/23418H04N21/2187

Abstract

Claims (72)

1. A computer-implemented method, comprising:

receiving media content including one or more transcoding parameters;

determining, from the one or more transcoding parameters, a set of performance affecting parameters associated with transcoding operations;

generating a transcoding signature corresponding to at least the set of performance affecting parameters;

comparing the transcoding signature against a set of historical transcoding signatures corresponding to historical sets of performance affecting parameters;

determining a difference between the transcoding signature and individual historical signatures in the set of historical signatures exceeds a threshold;

determining, based at least in part on a heuristic database, a processing configuration for the media content;

assigning the media content to a set of cores of one or more resources based, at least in part on the processing configuration;

executing a transcoding operation for the media content;

determining a performance record for the transcoding operation exceeds a performance threshold; and

adding the transcoding signature, with the processing configuration used for the transcoding operation, to the set of historical transcoding signatures.

2. The computer-implemented method of claim 1 , wherein the media content is a live stream.

3. The computer-implemented method of claim 1 , wherein the one or more performance affecting parameters include at least one of a codec, a resolution, or a bitrate.

4. The computer-implemented method of claim 1 , further comprising:

receiving second media content including one or more second transcoding parameters;

determining, from the one or more second transcoding parameters, a set of second performance affecting parameters;

generating a second transcoding signature associated with at least the set of second performance affecting parameters;

comparing the second transcoding signature against the set of historical signatures;

determining the second transcoding signature matches a historical signature of the set of historical signatures; and

assigning the second media content to a second set of cores of the one or more resources based, at least in part on a configuration of the historical signature.

5. The computer-implemented method of claim 1 , further comprising:

identifying a second set of cores of the one or more resources; and

migrating at least part of the transcoding operation to be processed on at least a portion of the second set of cores.

6. A computer-implemented method, comprising:

determining, from a channel and an associated signature, a set of performance affecting parameters;

determining, based on the set of performance affecting parameters and at least a portion of historical parameters for at least a portion of the performance affecting parameters, a testing configuration for the channel;

assigning the channel to one or more resources according to the testing configuration;

determining one or more additional settings, for a transcoding operation using the testing configuration, corresponding to one or more unknown properties of the channel;

executing the transcoding operation for the channel;

determining testing values for one or more metrics, of the one or more resources, are within a threshold; and

storing the testing configuration for the set of performance affecting parameters.

7. The computer-implemented method of claim 6 , further comprising:

determining second testing values for the one or more metrics exceed the threshold; and

adjusting one or more attributes of the testing configuration.

8. The computer-implemented method of claim 6 , wherein the one or more performance affecting parameters include at least one of a codec, a resolution, or a bitrate.

9. The computer-implemented method of claim 6 , further comprising:

receiving, from a heuristics database, historical transcoding data including at least the historical parameters and historical configurations;

training a machine learning system based, at least in part, on the historical transcoding data;

providing, to the trained machine learning system, the set of performance affecting parameters; and

determining the testing configuration.

10. The computer-implemented method of claim 6 , further comprising:

determining a number of cores to process the channel;

determining available cores for the one or more resources; and

assigning the channel to a set of cores of the available cores.

11. The computer-implemented method of claim 10 , wherein the set of cores are consecutive cores.

12. The computer-implemented method of claim 10 , wherein the set of cores are associated with at least one of a common socket, a common die, a common bus, or a shared memory cache.

13. The computer-implemented method of claim 10 , wherein at least one core of the set of cores is a fragmented core that shares at least a portion of core processing capability with at least one additional operation.

14. The computer-implemented method of claim 6 , further comprising:

determining available cores for the one or more resources;

assigning the channel to a set of cores of the available cores;

identifying a second set of the available cores; and

migrating at least a portion of the channel to at least one core of the second set of the available cores.

15. A system, comprising:

at least one processor; and

memory including instructions that, when executed by the at least one processor, cause the system to:

determine, from a channel and an associated signature, a set of performance affecting parameters;

determine, based on the set of performance affecting parameters and at least a portion of historical parameters for at least a portion of the performance affecting parameters, a testing configuration for the channel;

assign the channel to one or more resources according to the testing configuration;

determine one or more additional settings, for a transcoding operation using the testing configuration, corresponding to one or more unknown properties of the channel;

execute the transcoding operation for the channel;

determine testing values for one or more metrics, of the one or more resources, are within a threshold; and

store the testing configuration for the set of performance affecting parameters.

16. The system of claim 15 , wherein the instructions when executed further cause the system to:

determine second testing values for the one or more metrics exceed the threshold; and

adjust one or more attributes of the testing configuration.

17. The system of claim 15 , wherein the one or more performance affecting parameters include at least one of a codec, a resolution, or a bitrate.

18. The system of claim 15 , wherein the instructions when executed further cause the system to:

determine a number of cores to process the channel;

determine available cores for the one or more resources; and

assign the channel to a set of cores of the available cores.

19. The system of claim 18 , wherein the set of cores are associated with at least one of a common socket, a common die, a common bus, or a shared memory cache.

20. The system of claim 18 , wherein at least one core of the set of cores is a fragmented core that shares at least a portion of core processing capability with at least one additional operation.

Assignments (1)

References Cited (9)