Stackable Network Switch Failover via Loop Flag Rerouting
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Conventional failover solutions for stackable network switches result in long system downtime due to the involvement of control plane processors in reconfiguring stacking interfaces and routing traffic around failed links.
Innovation Solution
Implementing a stack unit with two stacking interfaces assigned as alternates, a forwarding engine that toggles a loop flag in packets to reroute them through the alternate interface when a stacking link fails, and using a device identifier to manage packet forwarding and dropping, ensuring packets are only transferred when the loop flag is not set or for unicast packets targeting the correct device.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Reliability
If control plane processors are used to reconfigure stacking interfaces during link failure, then routing around failed links is achieved, but system downtime increases
Solution Approach 1:
The patent segments the failover function from the control plane processor to the forwarding engine. The forwarding engine independently handles link failure detection and packet rerouting without requiring control plane processor intervention, thus achieving fast failover while maintaining reliability.
Solution Approach 2:
The patent implements preliminary action by pre-configuring alternate stacking interfaces and establishing backup forwarding paths before link failures occur. When a link fails, the forwarding engine can immediately switch to the pre-configured alternate interface without waiting for control plane reconfiguration.
2Adaptability or versatility
If stacking links are used to interconnect stack units, then network scalability is improved, but link failures can cause system disruption
Solution Approach 1:
The patent applies beforehand cushioning by implementing redundant alternate stacking interfaces and pre-established backup forwarding paths. When a stacking link fails, the system has pre-prepared alternative routes to cushion the impact, maintaining system stability while preserving the scalability benefits of stacking.
3Reliability
If conventional failover reconfiguration is performed, then traffic routing around failed links is achieved, but forwarding efficiency decreases due to control plane involvement
Solution Approach 1:
The patent extracts the failover and rerouting functions from the control plane processor and places them in the forwarding engine. This extraction allows traffic continuity to be maintained while preserving forwarding efficiency, as the forwarding engine can handle rerouting at line speed without control plane processing delays.
Data Source
AI summary
A network switch including a plurality of stack units sequentially linked to each other. Each stack unit includes a port, a forwarding engine, a first interface, and a second interface. The second interface of each stack unit is linked to the first interface of a subsequent stack unit. The second interface of a last stack unit is linked to the first interface of a first stack unit. In response to a packet traveling in a first direction from the first stack unit to the last stack unit and a link between a second stack unit and a third stack unit being inoperative, the forwarding engine of the second stack unit returns the packet in a second direction to a preceding stack unit. The packet travels in the second direction from the preceding stack unit to the first stack unit, to the last stack unit, and to the third stack unit.