Transport Accelerator Chunking for Live Streaming Latency
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Current content transfer techniques are inadequate for scenarios where a user device generates and transmits a content stream, such as live events or media broadcasts, as they are primarily designed for content consumption, leading to issues like inaccurate download rate estimation and increased rebuffering during live streaming.
Innovation Solution
A transport accelerator on a user device subdivides content fragments into multiple chunks and uses a connection manager to transmit these chunks via multiple connections established with a remote recipient, optimizing the transmission process to improve performance and reduce latency.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Productivity
If content is transmitted as complete fragments over a single connection, then transmission simplicity is maintained, but transmission speed and efficiency deteriorate
Solution Approach 1:
The transport accelerator subdivides content fragments into multiple smaller chunks and transmits them through multiple parallel connections simultaneously. This segmentation increases transmission speed by utilizing multiple network paths, while the transport accelerator manages the complexity of chunk assembly and connection coordination centrally.
2Reliability
If a single connection is used for content transmission, then connection management is simplified, but transmission reliability and throughput deteriorate
Solution Approach 1:
Content is divided into multiple chunks transmitted over multiple independent connections, providing redundancy and fault tolerance. If one connection fails, other connections can still deliver content, improving transmission reliability while the transport accelerator handles the complexity of managing multiple connections.
Solution Approach 2:
Multiple parallel connections are combined and coordinated by the transport accelerator to achieve higher overall throughput and reliability. The accelerator merges the data streams from multiple connections and reassembles the original content, providing both diversity gain and simplified end-point implementation.
3Measurement precision
If download rate estimation is based on sequential fragment reception, then estimation accuracy is improved for recorded content, but responsiveness to live content changes deteriorates
Solution Approach 1:
By dividing content into smaller chunks transmitted over multiple connections, the system can more rapidly sample and estimate download rates. The parallel transmission provides more frequent feedback points for rate estimation, improving both accuracy and responsiveness to changing network conditions in live streaming scenarios.
Data Source
AI summary
Transport accelerator (TA) systems and methods for accelerating transmission of content from a user agent (UA) of a user device to a remote recipient are provided according to embodiments of the present disclosure. Embodiments comprise a TA architecture implementing a connection manager (CM) and a request manager (RM). A RM of embodiments subdivides fragments of content provided by the UA into a plurality of content chunks, each fragment may be subdivided into multiple content chunks. The RM of embodiments provides content chunks to a connection manager (CM) of the TA for transmitting the content chunks. The CM of embodiments transmits the content chunks via a plurality of connections established between the CM and the remote recipient.