Edge Node Return Path Learning for Layer 2 Multipathing
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
In layer 2 networks, the increase in MAC addresses leads to flooding traffic and excessive network resource consumption, limiting the ability to implement multipathing and traffic engineering, even with MPLS technology.
Innovation Solution
A communication system with edge nodes that set and learn return paths based on pre-configured path selection information, reducing unnecessary packet duplication and enabling multipathing while optimizing network resource usage.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Adaptability or versatility
If MPLS is utilized for layer 2 virtual networking, then network routing capability is improved, but path selection and traffic engineering cannot be executed
Solution Approach 1:
The patent segments the routing function by introducing separate return path learning and selection mechanisms alongside the existing forward path configuration. Edge nodes independently learn return paths through flooded packets, enabling multipathing while maintaining layer 2 transparency. This segmentation allows both forward and return traffic to be engineered separately.
Solution Approach 2:
The patent implements preliminary action by pre-configuring multiple return path candidates at edge nodes before actual traffic flow. When flooded packets are received, edge nodes have already prepared multiple potential return paths, enabling rapid path selection without real-time computation delays.
2Adaptability or versatility
If MAC addresses are increased to support more bases, then network coverage is improved, but flooding traffic and resource consumption increase
Solution Approach 1:
The patent introduces return path information as an intermediary element carried within flooded packets. Instead of increasing MAC addresses for routing purposes, the system uses existing flooded packets to convey return path data, eliminating the need for additional addressing overhead while maintaining network coverage.
Solution Approach 2:
The patent changes the parameter being learned from simple MAC address mapping to structured return path information including multiple path candidates. This parameter transformation allows richer routing information to be captured without increasing the fundamental addressing scheme, reducing flooding traffic while improving routing capability.
3Adaptability or versatility
If return path information is learned through flooded packets, then multipathing is enabled, but path selection complexity increases
Solution Approach 1:
The patent implements self-service by enabling edge nodes to autonomously learn and select return paths without centralized control. Each edge node independently processes flooded packets, extracts return path information, and maintains its own path candidate table, distributing the complexity across multiple nodes rather than concentrating it in a single controller.
Solution Approach 2:
The patent reduces runtime complexity through preliminary action by pre-configuring multiple return path candidates before traffic flow begins. When packets arrive, edge nodes simply select from pre-prepared candidates based on stored information, avoiding complex real-time path computation and reducing operational complexity.
Data Source
AI summary
For a flooded packet transmitted from a first base to a second base connected to a transport network, a first edge node in the transport network first sets path selection information for selecting a return path for transmitting a packet from the second base to the first base, and then transmits the packet to the second base via the path configured in advance. When receiving the flooded packet, a second edge node selects the return path on the basis of the path selection information from a plurality of path candidates configured in advance between the first base and the second base and learns the path in association with information indicating the transmission source.