Spanning Tree Protocol Flush Optimization for Edge Port Traffic
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Current spanning tree protocols (STP) cause unnecessary traffic flooding on edge ports during topology changes, leading to link saturation and resource wastage, as they clear all forwarding table entries and relearn addresses, even on unaffected ports.
Innovation Solution
Disable address learning during a topology change period, update forwarding table entries to flood frames only on network ports, and re-enable learning after the change to repopulate entries efficiently, minimizing unnecessary traffic on edge ports.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Reliability
If all forwarding table entries are cleared and traffic is flooded on all ports during topology changes, then connectivity restoration is achieved, but link saturation and resource wastage occur on edge ports
Solution Approach 1:
The patent segments the flooding process by port type, separating network ports from edge ports. During topology changes, frames are flooded only on network ports while edge ports are excluded, thereby segmenting the flooding scope to avoid unnecessary resource consumption on edge ports while maintaining connectivity restoration through network ports.
Solution Approach 2:
The patent applies different quality treatments to different port types. Network ports receive full flooding treatment to ensure connectivity restoration, while edge ports are selectively excluded from flooding to prevent resource wastage. This local differentiation resolves the contradiction by optimizing resource allocation based on port function.
2Reliability
If topology change messages are repeated to ensure reliability, then message delivery is guaranteed, but newly learned address entries are flushed again causing stress and resource waste
Solution Approach 1:
The patent implements preliminary filtering of topology change messages before they trigger address entry flushing. By checking whether address learning is currently disabled, the system prevents redundant flushing operations that would occur due to repeated topology change messages, thereby protecting newly learned entries and improving resource efficiency while maintaining reliable message delivery.
Solution Approach 2:
The patent introduces a feedback mechanism where the state of address learning (enabled/disabled) is checked before executing the flush operation. This feedback loop prevents unnecessary flushing when address learning is already disabled, eliminating the wasteful repetition caused by repeated topology change messages and improving overall system productivity.
Data Source
AI summary
In one embodiment, in response to receiving a topology change notification at a network bridge having ports identified as either a network port or an edge port, address learning may be disabled on the network bridge. Once address learning is disabled, an association of all entries of a forwarding table of the network bridge having addresses previously forwarded on a particular network port of the network bridge may be changed to forward those addresses on all network ports of the network bridge (e.g., flooding the frames not addressed to edge ports on all network ports only). Subsequently, address learning may be enabled on the network bridge, thus repopulating the network port entries of the forwarding table in response to the topology change.