Edge Location Server for Streaming Content Routing

Resolve Bottlenecks,
Find Innovative SolutionsGenerate Solutions

Solution Overview

Problem

Current telecommunications systems face challenges in efficiently delivering streaming content due to variability in user demand, leading to suboptimal content availability and response times, as they lack a method to accurately predict and prioritize content distribution based on aggregated user viewing data.

Innovation Solution

An edge location server determines relative demand profiles for media content items using aggregated viewing information from a streaming application, prioritizes content, and directs content servers to ensure availability at accessible edge servers, providing location information for rapid content access.

Engineering Contradictions & Design Principles

VSEngineering Contradiction Analysis

1Reliability

If content is distributed uniformly across all edge servers, then content availability is maintained, but response time increases and system efficiency deteriorates

Engineering Contradiction:

Improvecontent availabilityVSAvoidresponse time

Core Design Contradiction:

ReliabilityVSLoss of time

Solution Approach 1:

The system performs preliminary actions by predicting future content demand using machine learning models before users actually request the content. Edge servers pre-cache predicted high-demand content items based on aggregated viewing data, historical patterns, and real-time analytics, ensuring content is ready for immediate delivery when requested, thus reducing response time while maintaining availability

Inventive Principle:

Principle #10Preliminary action

Solution Approach 2:

The system applies local quality by distributing content selectively to specific edge servers based on local demand characteristics. Each edge server receives and caches content tailored to its geographic region's viewing patterns, rather than uniform distribution. This localized content placement optimizes response time for each region while maintaining overall system reliability

Inventive Principle:

Principle #3Local quality

2Productivity

If machine learning based routing is implemented, then content delivery efficiency is improved, but device complexity and computational requirements increase

Engineering Contradiction:

Improvecontent delivery efficiencyVSAvoidsystem complexity

Core Design Contradiction:

ProductivityVSDevice complexity

Solution Approach 1:

The system segments the machine learning workload across multiple components: centralized training servers handle model training, edge location servers handle inference and routing decisions, and content servers handle delivery. This segmentation distributes computational complexity, allowing each component to be optimized independently and reducing the complexity burden on any single device while maintaining high delivery efficiency

Inventive Principle:

Principle #1Segmentation

Solution Approach 2:

The system introduces machine learning models as intermediaries between content requests and delivery. These models act as intelligent mediators that analyze aggregated viewing data, predict demand, and generate routing decisions, thereby automating complex content placement and delivery optimization without requiring direct complex control logic in each system component

Inventive Principle:

Principle #24Intermediary (Mediator)

3Loss of time

If content is cached at edge servers, then response time is reduced, but network bandwidth consumption and cache management complexity increase

Engineering Contradiction:

Improveresponse timeVSAvoidnetwork bandwidth consumption

Core Design Contradiction:

Loss of timeVSLoss of energy

Solution Approach 1:

The system applies partial action by caching only a subset of content at edge servers - specifically, content predicted to have high demand in each region. Rather than caching all content, the machine learning models identify and pre-cache only the most relevant content items, reducing unnecessary bandwidth consumption while still achieving fast response times for predicted popular content

Inventive Principle:

Principle #16Partial or excessive action

Solution Approach 2:

The system implements feedback mechanisms where aggregated viewing data from users is continuously collected and fed back to the machine learning models. This feedback loop enables dynamic adjustment of cache content based on actual viewing patterns, ensuring that cached content remains relevant and in-demand, thereby optimizing the trade-off between response time and bandwidth usage

Inventive Principle:

Principle #23Feedback

Data Source

PatentUS11558649B2Method and an apparatus for controlling content delivery via machine-learning based routing

Publication Date: 2023.01.17 AT&T INTELLECTUAL PROPERTY I L P

AI summary

Aspects of the subject disclosure may include, for example, receiving, from a streaming application, a listing of media content items provided by the streaming application to a first device of a first user, determining a priority set of the media content items of the listing of media content items according to relative demand profiles of a plurality of media content items, for each priority media content item of the priority set of the media content items, providing the priority media content item to an edge server of a set of edge servers accessible to the first device, updating edge server location information associated with the priority media content item, and providing the edge server location information to the streaming application. Other embodiments are disclosed.