Cloud Media Transcoding Through Splitter-Merger Parallel Processing
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing media transcoding processes are inefficient and slow due to the limitations of single-stream transcoding, which can be addressed by implementing parallel transcoding techniques using multiple transcoders.
Innovation Solution
A method and system for parallel transcoding of media content by splitting a compressed video stream into multiple sub-streams, transcoding each sub-stream in parallel using multiple transcoders, and merging them back into a single transcoded stream, with real-time reporting and management of the transcoding process.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Productivity
If single-stream transcoding is used, then device complexity is low, but productivity is poor
Solution Approach 1:
The patent divides a single video stream into multiple parallel sub-streams, each processed by separate transcoder instances. This segmentation enables concurrent processing of different portions of the content, significantly increasing overall transcoding throughput while maintaining manageable complexity through modular architecture
Solution Approach 2:
Multiple parallel transcoding operations are merged into a unified workflow managed by the NBMP workflow manager. The merger task combines results from multiple transcoder tasks, allowing the system to achieve high productivity through parallel processing while presenting a single integrated interface to users
2Productivity
If parallel transcoding with multiple transcoders is implemented, then productivity is improved, but device complexity increases
Solution Approach 1:
The NBMP workflow manager implements a universal control mechanism that handles multiple transcoder tasks, splitter operations, and merger functions through a single standardized interface. This multi-functional approach manages the complexity of parallel processing while maintaining system coherence and enabling scalability
Solution Approach 2:
The system implements comprehensive feedback mechanisms where media processing entities report their states (splitter state, transcoder state, merger state) back to the workflow manager. This feedback enables real-time monitoring and coordination of parallel operations, managing complexity through automated control loops
3Measurement precision
If real-time reporting of processing states is implemented, then measurement precision is improved, but use of energy increases
Solution Approach 1:
The system implements selective reporting where media processing entities report only the necessary state information (splitter state, transcoder state, or merger state) required for workflow coordination. This partial reporting approach provides sufficient measurement precision for monitoring while minimizing unnecessary communication overhead and energy consumption
Data Source
Figure 1
Figure 2
Figure 3
AI summary
Systems and methods for parallel transcoding of media content are provided, where the media content is split into parallel media sub-streams and each media sub-stream is transcoded using one transcoder and then the parallel media sub-streams are merged into a single transcoded stream. The parallel transcoding may be performed by one or more media processing entities as a part of a workflow that includes a splitter task, transcoder tasks, and a merger task, and the one or more media processing entities may be controlled to report states of the tasks.