Smart Delivery Node Neural Network Traffic Categorization

Resolve Bottlenecks,
Find Innovative Solutions
Generate Solutions

Solution Overview

Problem

Content delivery networks (CDNs) face performance degradation when handling a mixture of different types of traffic, as they are typically optimized for video traffic, leading to inefficiencies and increased bandwidth usage on origin servers.

Innovation Solution

Implementing a method within delivery nodes that uses a trained neural network to categorize content and redirect requests to the most suitable delivery nodes based on traffic patterns, optimizing access policies and improving resource allocation across the CDN.

Engineering Contradictions & Design Principles

VSEngineering Contradiction Analysis

1Productivity

If delivery nodes are optimized for video traffic, then video delivery performance is improved, but performance degrades when handling mixed traffic types

Engineering Contradiction:
Improvevideo delivery performanceVSAvoidhandling mixed traffic types
Core Design Contradiction:
ProductivityVSAdaptability or versatility

Solution Approach 1:

The patent segments traffic into different categories (video traffic and non-video traffic) and routes them to different delivery nodes. Delivery nodes are specialized to handle specific traffic types, with video-optimized nodes for video content and general-purpose nodes for other content types. This segmentation resolves the contradiction by allowing each node type to be optimized for its specific function while the system as a whole handles diverse traffic types.

Inventive Principle:
Principle #1Segmentation

2Productivity

If delivery nodes cache and deliver small data files for web traffic, then web traffic handling is improved, but storage I/O performance deteriorates

Engineering Contradiction:
Improveweb traffic handlingVSAvoidstorage I/O operations
Core Design Contradiction:
ProductivityVSLoss of energy

Solution Approach 1:

The patent applies local quality by creating delivery nodes with different characteristics suited to different traffic types. General-purpose delivery nodes are configured with storage and I/O optimizations appropriate for web traffic patterns, while video-optimized nodes have different configurations. This allows each node type to have local optimizations that reduce I/O overhead for its specific traffic type while maintaining overall system versatility.

Inventive Principle:
Principle #3Local quality

3Adaptability or versatility

If delivery nodes handle both video and web traffic, then CDN versatility is improved, but overall performance degrades

Engineering Contradiction:
Improvetraffic type coverageVSAvoidoverall CDN performance
Core Design Contradiction:
Adaptability or versatilityVSProductivity

Solution Approach 1:

The patent introduces a request router as an intermediary component that receives client requests, categorizes the traffic type, and routes requests to appropriate specialized delivery nodes. This intermediary layer enables the CDN to handle diverse traffic types effectively by directing each request to the most suitable node, thereby maintaining high overall performance while preserving system versatility.

Inventive Principle:
Principle #24Intermediary (Mediator)

Data Source

PatentUS11350145B2Smart delivery node
Publication Date: 2022.05.31 TELEFONAKTIEBOLAGET LM ERICSSON (PUBL)
  • US11350145B2 patent drawing
  • US11350145B2 patent drawing
  • US11350145B2 patent drawing

AI summary

A method, Delivery Node, DN, and Content Delivery Network, CDN, are optimized to deliver content of different categories. The DN is operative to receive a request for a content, obtain a determination of whether or not a category of content is associated with the requested content and responsive to the determination that no category of content is associated with the requested content, forward the request for the content towards an origin server and upon receiving a response from the origin server serve the content. The CDN comprises a plurality of DNs, a data processing service operative to obtain, from the DNs, and assemble, data sets into training and validation data formatted to be used for training and validating a Neural Network, NN. The CDN comprises a NN training service, operative to train and validate the NN and a configuration service, operative to configure the plurality of DNs with the trained and validated NN.