Network Access Cluster Routing for Unequal Uplink Failures
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
In a network access cluster with unequal uplink bandwidth, traffic flow inefficiencies and oversubscription occur when an uplink fails, leading to dropped packets and failed transmissions due to the lack of immediate alternative routing paths.
Innovation Solution
Implement a mechanism where a network device with lower bandwidth sets the link cost to zero upon detecting an uplink failure, allowing traffic to be rerouted through a peer device with higher bandwidth, and revert to the original link cost when bandwidth equality is restored, preventing loops.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Device complexity
If traffic is forwarded only via remaining uplinks from node A when an uplink fails, then the routing path is simple and direct, but traffic flow efficiency deteriorates due to oversubscription and lack of alternative paths
Solution Approach 1:
The patent implements dynamic link cost adjustment based on real-time uplink status. When an uplink fails, the system dynamically changes the cost of the peer link from its original value to zero, enabling automatic traffic redistribution. This dynamic parameter adjustment resolves the contradiction by adapting the routing complexity to match current network conditions, improving traffic flow efficiency without permanent complexity increase.
Solution Approach 2:
The system changes the cost parameter of the peer link dynamically. When node A detects uplink failure, it modifies the link cost parameter to zero, making the peer link through node B an attractive alternative path. This parameter change enables the network to switch from a simple direct-routing model to an optimized multi-path model, resolving the efficiency-contradiction.
2Productivity
If the link cost is set to zero to enable alternative routing paths, then traffic load-sharing improves, but routing loop risk increases
Solution Approach 1:
The patent implements a feedback mechanism where nodes continuously monitor uplink status and dynamically adjust link costs accordingly. When an uplink fails, the peer link cost is set to zero; when the uplink recovers, the cost returns to its original value. This feedback-based dynamic adjustment prevents routing loops by ensuring that zero-cost paths only exist when genuinely needed, resolving the contradiction between improved load-sharing and loop prevention.
Solution Approach 2:
The link cost parameter is made dynamic rather than static, changing from its original value to zero and back again based on uplink status. This temporal dimension to the routing parameters allows the network to temporarily utilize alternative paths when necessary while automatically preventing loops when the original uplinks are restored, balancing load-sharing improvement with loop prevention.
3Device complexity
If all uplinks from node A must fail before traffic flows via node B, then routing decisions are simple, but packet loss increases during partial failures
Solution Approach 1:
The patent prepares alternative routing paths in advance by establishing peer links between cluster nodes and configuring dynamic cost adjustment mechanisms. When uplink failures occur, these pre-prepared paths can be immediately activated by setting the peer link cost to zero, without requiring complex real-time routing calculations. This preliminary preparation resolves the contradiction by having simple activation logic ready in advance.
Solution Approach 2:
The patent introduces the peer link as an intermediary path between node A and the upstream network. Instead of requiring direct uplinks to be completely failed before alternative routing, the peer link serves as a mediator that can carry traffic through node B when node A's uplinks are degraded. This intermediary mechanism improves reliability during partial failures while maintaining simple routing decisions through automatic cost-based selection.
Data Source
AI summary
A system transmits, by a network access cluster, data to an upstream network device via a plurality of uplinks. The network access cluster comprises a first and a second network device, which each communicate with the upstream network device via a first and second group of uplinks. The first and second network devices communicate with each other via a link. The system updates a first bandwidth associated with the first network device in response to detecting a failure in the first group of uplinks. Responsive to the first bandwidth being less than a second bandwidth associated with the second network device, the system sets a forwarding cost (e.g., to zero) of the link. Based on the zero-cost link, the system allows an additional path via the second network device for transmitting data received by the first network device to the upstream network device.