Distributed Routing Nodes for Cloud Routing and Packet Scheduling
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing network devices, such as commercial edge routers and internal routers, face challenges in large-scale Internet routing with limited options, high costs, poor function scalability, and slow iteration speed, making them unsuitable for cloud scenarios.
Innovation Solution
A distributed system architecture comprising first and second routing node devices and a global routing service node device, connected through a backbone network, maintains a mapping relationship between peering connections and scheduling labels, enabling efficient transmission scheduling of service packets based on this mapping.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Adaptability or versatility
If commercial routers are used for large-scale Internet routing, then routing function is provided, but device complexity and cost increase significantly
Solution Approach 1:
The patent segments the routing function from the forwarding function, creating independent routing nodes that only perform routing decisions while universal switching components handle packet forwarding. This separation allows routing logic to be simplified and distributed across multiple nodes rather than concentrated in complex commercial routers.
Solution Approach 2:
The patent introduces universal switching components that can perform multiple functions including packet forwarding, scheduling, and routing information storage. These universal components replace specialized commercial router hardware, providing the same routing capabilities with simpler, more versatile equipment.
2Adaptability or versatility
If commercial routers are used for large-scale Internet routing, then routing function is provided, but cost increases significantly
Solution Approach 1:
The patent uses software-based routing nodes that can be replicated and distributed across multiple standard servers rather than requiring expensive proprietary commercial router hardware. The routing functionality is copied across numerous low-cost nodes instead of being implemented in a few high-cost devices.
Solution Approach 2:
The patent employs standard off-the-shelf servers and universal switching components instead of expensive commercial router equipment. These cheaper components can be easily replaced and upgraded, reducing both initial investment and long-term operational costs.
3Measurement precision
If full routing information is maintained in distributed system, then routing accuracy is improved, but transmission overhead increases
Solution Approach 1:
The patent extracts only the essential routing information needed for forwarding decisions and stores it locally in universal switching components. Full routing tables are not transmitted across the network; instead, only necessary forwarding entries are cached at each node, reducing transmission overhead while maintaining routing accuracy.
Solution Approach 2:
The patent introduces a centralized routing information management system that acts as an intermediary, distributing only necessary routing entries to appropriate nodes rather than transmitting complete routing tables. This mediator optimizes information distribution by filtering and directing only relevant data to each node.
Data Source
AI summary
A distributed system includes a first routing node device, a second routing node device, and a global routing service node device, connected to a backbone network. The first routing node device is distributed in an access network and configured to access one or more Internet service provider (ISP) networks, and includes a universal service component and a first universal switching component. The second routing node device is distributed in a data center network and configured to provide data connection between network modules in the data center network, and includes a second universal switching component. The global routing service node device is arranged outside the access network and the data center network. A mapping relationship between peering connections and scheduling labels is maintained in the distributed system and transmitted between node devices. The node devices perform transmission scheduling of service packet in the cloud network based on the mapping relationship.