Method and system for a graph based video streaming platform

• US 9,417,921 B2
• Filed: 07/31/2014
• Issued: 08/16/2016
• Est. Priority Date: 07/31/2014
• Status: Active Grant

• Current Assignee: iStreamPlanet Co. LLC (Warner Bros. Discovery, Inc.)
• Original Assignee: iStreamPlanet
• Inventors: Galdy, Alessio, Sapru, Sidharth
• Primary Examiner(s): Khatri, Anil
• Application Number:
  US14/448,915
• Publication Number:
  US 20160034306A1
• Time in Patent Office: 747 Days
• Field of Search: 718/1, 718101-107, 718/108
• US Class Current: 1/1
• CPC Class Codes: G06F 9/4881 Scheduling strategies for dispatcher, e.g. round robin, multi-level priority queues
• CPC Class Codes: G06F 9/5011 the resources being hardware resources other than CPUs, Servers and Terminals
• CPC Class Codes: G06F 9/5066 Algorithms for mapping a plurality of inter-dependent sub-tasks onto a plurality of physical CPUs mappping at compile time, see G06F8/451

Abstract

A method implemented in an electronic device serving as an orchestrator managing video and audio stream processing of a streaming platform system is disclosed. The method includes the electronic device receiving a request to process a video source and creating a task graph based on the request, where the task graph is a directed acyclic graph of tasks for processing the video source, where each node of the task graph represents a processing task, and where each edge of the task graph represents a data flow across two processing tasks and corresponding input and output of each processing task. The method also includes the electronic device estimating resource requirements of each processing tasks, and splitting the task graph into a plurality of subsets, wherein each subset corresponds to a task group to be executed by one or more workers of a plurality of processing units of the streaming platform system.

Claims

1. A method implemented in an electronic device serving as an orchestrator managing video and audio stream processing of a streaming platform system, the method comprising:

receiving a request to process a video source, the request containing parameters describing the video source;

creating a task graph based on the request, wherein the task graph is a directed acyclic graph of tasks for processing the video source, wherein each node of the task graph represents a processing task, and wherein each edge of the task graph represents a data flow across two processing tasks and corresponding input and output of each processing task;

estimating resource requirements of each processing task, where estimating the resource requirement includes any one of estimating storage usage of each processing task during its execution cycle, estimating communication bandwidth usage of each processing task during its execution cycle, or estimating computing resource consumption of each processing task during its execution cycle; and

splitting the task graph into a plurality of subsets, wherein each subset corresponds to a task group to be executed by one or more workers of a plurality of workers, where each worker is a processing unit of the streaming platform system.

2. The method of claim 1, the method further comprising:

after estimating the resource requirement of each processing task, revising the task graph prior to splitting the task graph into the plurality of subsets based on at least one of;

the estimated resource requirement of one or more processing tasks; and

redundancy requirements of one or more processing tasks.

3. The method of claim 2, the method further comprising:

after revising the task graph and after splitting the task graph into the plurality of subsets, estimating the resource requirement of each processing task again to accommodate any change as of resource requirement as a result of the revision.

4. The method of claim 1, the method further comprising:

storing a status of each subset of the task graph; and

updating the status of each subset when the subset is allocated to one or more workers for execution.

5. The method of claim 1, wherein the parameters describing the video source include at least one of:

a type of the video source;

an address of the video source;

an indication of whether to enable caption or enable advertisement insertion process;

desired video and audio transcoding operations by the streaming platform system;

desired content protection operations by the streaming platform system; and

desired publishing operations to output from the streaming platform system.

6. The method of claim 1, wherein estimating communication bandwidth usage and computing resource consumption of each processing task include predictions based on the video resource being ingested in the streaming platform system.

7. The method of claim 1, wherein each subset corresponds to a task group to be executed by a single worker of a plurality of workers, and wherein the task graph contains a single input task.

8. An electronic device to serve as an orchestrator managing video and audio streams of a streaming platform system, the electronic device comprising:

a non-transitory machine-readable storage medium to store the orchestrator; and

a processor coupled to the non-transitory machine-readable storage medium, the processor to execute the orchestrator, the orchestrator configured to receive a request to process a video source, the request containing parameters describing the video source, create a task graph based on the request, wherein the task graph is a directed acyclic graph of tasks for processing the video source, wherein each node of the task graph represents a processing task, and wherein each edge of the task graph represents a data flow across two processing tasks and corresponding input and output of each processing task, estimate resource requirements of each processing task, where the estimate of the resource requirements include any one of an estimate of storage usage of each processing task during its execution cycle, an estimate of communication bandwidth usage of each processing task during its execution cycle, or an estimate of computing resource consumption of each processing task during its execution cycle, and split the task graph into a plurality of subsets, wherein each subset corresponds to a task group to be executed by one or more workers of a plurality of workers, where each worker is a processing unit of the streaming platform system.

9. The electronic device of claim 8, wherein the orchestrator is further configured to:

after estimating the resource requirement of each processing task, revise the task graph prior to splitting the task graph into the plurality of subsets based on at least one of;

the estimated resource requirement of one or more processing tasks; and

redundancy requirements of one or more processing tasks.

10. The electronic device of claim 8, wherein the orchestrator is further configured to:

store a status of each subset of the task graph; and

update the status of each subset when the subset is allocated to one or more workers for execution.

11. The electronic device of claim 8, wherein the parameters describing the video source include at least one of:

a type of the video source;

an address of the video source;

an indication of whether to enable caption or enable advertisement insertion process;

desired video and audio transcoding operations by the streaming platform system;

desired content protection operations by the streaming platform system; and

desired publishing operations to output from the streaming platform system.

12. The electronic device of claim 8, wherein each subset corresponds to a task group to be executed by a single worker of a plurality of workers, and wherein the task graph contains a single input task.

13. A non-transitory machine-readable medium having instructions stored therein, which when executed by a processor, cause the processor to perform operations in an electronic device serving as an orchestrator managing video and audio stream processing of a streaming platform system, the operations comprising:

receiving a request to process a video source, the request containing parameters describing the video source;

creating a task graph based on the request, wherein the task graph is a directed acyclic graph of tasks for processing the video source, wherein each node of the task graph represents a processing task, and wherein each edge of the task graph represents a data flow across two processing tasks and corresponding input and output of each processing task;

estimating resource requirements of each processing task, where estimating the resource requirements include any one of estimating storage usage of each processing task during its execution cycle, estimating communication bandwidth usage of each processing task during its execution cycle, or estimating computing resource consumption of each processing task during its execution cycle; and

splitting the task graph into a plurality of subsets, wherein each subset corresponds to a task group to be executed by one or more workers of a plurality of workers, where each worker is a processing unit of the streaming platform system.

14. The non-transitory machine-readable medium of claim 13, the operations further comprising:

after estimating the resource requirement of each processing task, revising the task graph prior to splitting the task graph into the plurality of subsets based on at least one of;

the estimated resource requirement of one or more processing tasks; and

redundancy requirements of one or more processing tasks.

15. The non-transitory machine-readable medium of claim 13, the operations further comprising:

storing a status of each subset of the task graph; and

updating the status of each subset when the subset is allocated to one or more workers for execution.

16. The non-transitory machine-readable medium of claim 13, the operations further comprising:

a type of the video source;

an address of the video source;

an indication of whether to enable caption or enable advertisement insertion process;

desired video and audio transcoding operations by the streaming platform system;

desired content protection operations by the streaming platform system; and

desired publishing operations to output from the streaming platform system.

17. The non-transitory machine-readable medium of claim 13, wherein each subset corresponds to a task group to be executed by a single worker of a plurality of workers, and wherein the task graph contains a single input task.