Hierarchical Internet Architecture Reducing Network Latency
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Traditional Internet network architectures face bottlenecks in user access speed, management complexity, and poor scalability due to their tree-like layering and stage-by-stage aggregation, which limits network efficiency and accessibility.
Innovation Solution
The proposed Internet system architecture introduces a scale-free network architecture with a reduced connection hop count and a network feedback mechanism, dynamically allocating resources to super nodes, allowing for efficient network management and fast access while reducing latency.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Ease of operation
If a traditional tree-like layering network architecture is used, then network management and routing are standardized, but user access speed decreases and network scalability becomes poor due to many network layers and stages
Solution Approach 1:
The patent segments the traditional monolithic network architecture into a hierarchical structure with a core network layer and multiple access network layers. This segmentation reduces the number of stages data must traverse by handling inter-domain routing at the core layer and local access at the edge layers, thereby improving user access speed while maintaining manageable network operations through layered responsibility.
Solution Approach 2:
The patent introduces a new dimensional organization to the network architecture by creating a multi-layer hierarchical structure that adds vertical stratification (core vs. access layers) to the traditional horizontal routing model. This dimensional change allows data to be routed more efficiently by leveraging the hierarchical structure, reducing the number of hops and improving access speed without sacrificing management standardization.
2Ease of operation
If a centralized network controller is deployed to control all OpenFlow switches, then domain control management is achieved, but network scalability deteriorates and processing efficiency decreases as network scale increases
Solution Approach 1:
The patent segments the centralized control function into distributed control entities deployed at different hierarchical levels (core network layer and access network layers). This segmentation allows each control entity to manage only its local domain, eliminating the single-point bottleneck of centralized control and enabling the network to scale horizontally by adding more distributed control entities without increasing central controller load.
Solution Approach 2:
The patent transforms the control architecture from a single-dimensional centralized model to a multi-dimensional distributed hierarchical model. Control functions are distributed across multiple dimensions (different layers and domains), allowing each control entity to operate independently within its scope while maintaining overall network coherence through standardized inter-layer communication protocols.
3Area of stationary object
If multiple network layers and stages are used for aggregation, then comprehensive network coverage is achieved, but network complexity increases and management becomes complex
Solution Approach 1:
The patent segments the network into distinct functional layers (core network layer and access network layers) with clearly defined responsibilities. The core layer handles wide-area routing and inter-domain traffic, while access layers handle local access and edge routing. This segmentation maintains comprehensive network coverage by preserving all necessary network functions while reducing overall complexity through functional separation and specialized processing at each layer.
Data Source
Figure 1~2
Figure 3
Figure 4~5
AI summary
The present invention discloses a network communication method, a device, and an Internet system. The method includes: receiving, by a first primary node at a first network layer, first communication information sent, by a non-primary node that initiates communication, to a non-primary node that receives communication, where the non-primary node that initiates communication is in a domain to which the first primary node belongs, and the non-primary node that receives communication is in a different domain at the first network layer; determining, by the first primary node, first label information, where the first label information is used to indicate a communication path, at a second network layer, from a node that receives the first communication information to a node that has a mapping relationship with a second primary node to which the non-primary node that receives communication belongs; and sending, by the first primary node, first information to a node at the second network layer, so that the node at the second network layer sends the first information to the second primary node according to the first label information by using the communication path corresponding to the first label information, where the first information includes the first communication information and the first label information.