Tune-In Segment for Low Latency Streaming
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Current low latency live streaming technologies face challenges with high transaction rates on Content Delivery Networks (CDNs) due to frequent segment requests and inefficient tune-in mechanisms, leading to increased latency and cacheability issues when clients join mid-segment, resulting in outdated data and increased CDN load.
Innovation Solution
A method for operating a media player and media providing entity that allows for a special tune-in segment to start with the desired fragment, rather than the first fragment, reducing the number of fragments in the segment and enabling efficient HTTP chunked delivery, allowing clients to tune-in closer to the live edge with reduced latency and cacheable responses.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Loss of time
If clients request segments starting from the first fragment, then the segment structure remains simple and cacheable, but clients experience increased latency when tuning in mid-segment
Solution Approach 1:
The segment is divided into multiple fragments, and a special tune-in segment is created that contains only the relevant fragments (e.g., fragments 3-10) starting from the desired tune-in point. This segmentation allows clients to skip unnecessary early fragments and start playing from the latest available content, reducing tune-in latency without complicating the overall segment structure.
Solution Approach 2:
The server pre-identifies the desired fragment number based on client request parameters and constructs the tune-in segment accordingly before transmission. This preliminary action of determining the start fragment and creating a truncated segment structure enables fast tune-in while maintaining simple HTTP delivery mechanisms.
2Productivity
If the server sends complete segments from the first fragment, then caching is simplified, but CDN load increases due to clients receiving outdated data
Solution Approach 1:
The server extracts and removes the unnecessary early fragments (e.g., fragments 1-2) from the segment, sending only the relevant later fragments (e.g., fragments 3-10) that contain fresh content. This extraction reduces the amount of data transmitted through the CDN, decreases CDN load, and ensures clients receive only the most current available content, improving both efficiency and data freshness.
3Speed
If segments are delivered with chunked encoding, then low latency is achieved, but transaction rates on CDN increase due to frequent segment requests
Solution Approach 1:
Instead of delivering complete segments from the first fragment, the server delivers a partial segment containing only the necessary fragments from the desired start point. This partial delivery reduces the amount of data that needs to be transmitted and cached, decreasing CDN transaction rates while maintaining low latency streaming performance.
Data Source
AI summary
The invention relates to a method for operating a media player when receiving a live stream which comprising a sequence of media segments, each media segment comprising a plurality of media fragments. The method comprisesdetermining a tune-in segment and a tune-in fragment of the live stream, at which the media player intends to start consuming the live stream,transmitting a media request towards a media providing entity, the media request comprising an indication indicating that the tune-in segment should start with the tune-in fragment as first media fragment,receiving a media response originating from the media providing entity, the media response comprising the tune-in segment which starts with the tune-in fragment as first media fragment.