Live Encoding for Long Tail Media Content Delivery
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Video on demand (VOD) services face high costs and resource inefficiencies in processing and storing large libraries of video content, particularly for less popular programs, leading to these being unavailable online due to the expense of encoding, packaging, and delivering multiple copies of varying quality.
Innovation Solution
Implementing a system that uses live encoding to create media streams on demand, reducing the need for pre-encoded copies of less popular content, thereby conserving processing, storage, and delivery resources, while maintaining efficient delivery of popular content through priori encoding and storage.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Speed
If multiple copies of video content are pre-encoded and stored for VOD delivery, then delivery speed and quality adaptation are improved, but processing costs and storage requirements increase significantly
Solution Approach 1:
The system performs preliminary encoding only for popular content that is likely to be requested, while leaving less popular content in its original format. This selective preliminary action reduces overall storage requirements while maintaining fast delivery for high-demand content.
Solution Approach 2:
Instead of creating multiple pre-encoded copies of all content, the system creates copies only when needed for popular content. For less popular content, it uses on-demand encoding to generate copies only when a user requests them, significantly reducing storage requirements.
2Adaptability or versatility
If video content is encoded multiple times to create varying quality versions, then adaptability to different viewing conditions is improved, but processing time and computational resources increase
Solution Approach 1:
The system performs preliminary encoding into multiple quality versions only for popular content that will be frequently accessed. For less popular content, it performs encoding on-demand only when needed, significantly reducing total processing time while maintaining quality adaptation capability.
Solution Approach 2:
The system dynamically adjusts its encoding strategy based on content popularity and request patterns. Frequently requested content receives pre-encoded multiple quality versions, while less popular content uses on-demand encoding, optimizing the balance between adaptability and processing time.
3Adaptability or versatility
If a large library of video content is made available for on-demand viewing, then service variety and customer interest are improved, but encoding and storage costs increase prohibitively
Solution Approach 1:
The system performs preliminary encoding and storage only for popular content in the large library, while keeping less popular content in original format or unencoded state. This selective preliminary action enables service variety while controlling storage costs.
Solution Approach 2:
The system creates encoded copies of content only when needed - either in advance for popular content or on-demand for less popular content. This approach allows the service to offer a large variety of content without incurring prohibitive storage costs for all possible encodings of all content.
4Productivity
If content is pre-encoded and stored before delivery, then delivery efficiency is improved, but up-front processing costs and computational resources increase
Solution Approach 1:
The system performs preliminary encoding and preparation only for content that is predicted to be popular or frequently requested. This selective preliminary action improves delivery efficiency for high-demand content while avoiding unnecessary up-front processing time for less popular content.
Solution Approach 2:
The system dynamically adjusts its pre-processing strategy based on content metadata, historical request patterns, and popularity predictions. This dynamic approach optimizes the balance between delivery efficiency and up-front processing time by pre-processing only the most beneficial content.
Data Source
AI summary
Systems, methods and devices are described to deliver media programs to remotely-located media players via a network. A media server receives a request for the media program from the remotely-located media player via the network and determines whether the requested media program has been previously encoded. If so, the previously-encoded media program is delivered to the remotely-located media player. If the requested media program has not been previously encoded, the requested media program is live encoded to thereby create a media stream encoding the requested media program in response to the request received from the remotely-located media player.

