Hierarchical Routing Topology for Resilient Network Path Availability
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing routing protocols face challenges in quickly responding to link failures without causing data traffic loss, as they require computation time for route recalculation, and existing solutions like Loop Free Alternates and Fast Local Rerouting are insufficient for all network nodes.
Innovation Solution
A hierarchical routing topology is created in a computing network with a single parent supernode and multiple child supernodes, allowing data packets to traverse along any available data link within a child supernode independently of established routing topologies, and exit network devices forward packets to the parent supernode for delivery, optimizing route establishment and avoiding loop formation.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Reliability
If existing routing protocols recalculate routes in response to detected failure, then route accuracy is maintained, but data traffic loss occurs due to computation time
Solution Approach 1:
The patent pre-calculates and stores multiple alternate paths to destination nodes before any failure occurs. When a link failure is detected, the routing protocol can immediately switch to a pre-computed alternate path without performing recalculation, thus maintaining route accuracy while eliminating computation time and associated data traffic loss.
Solution Approach 2:
The patent prepares backup routing paths in advance as a cushion against potential failures. These pre-established alternate paths act as a safety buffer that can be activated immediately upon failure detection, preventing the harmful effect of data traffic loss that would otherwise occur during route recalculation.
2Speed
If Loop Free Alternates protocol identifies feasible successor, then response speed to link failure improves, but coverage is limited to specific network nodes
Solution Approach 1:
The patent extends the fast rerouting capability of Loop Free Alternates to all network nodes by implementing a universal mechanism where every router pre-calculates and maintains multiple alternate paths to all destination nodes. This universal approach eliminates the limitations of LFA's selective node coverage while preserving the fast response speed advantage.
Solution Approach 2:
The patent moves from the localized, node-specific alternate path calculation of LFA to a comprehensive, network-wide approach where multiple alternate paths are maintained across all nodes. This dimensional expansion from point-specific to network-wide coverage enables both fast response and universal adaptability.
3Reliability
If Fast Local Rerouting performs complete recalculation, then route optimality is maintained, but computation time increases causing data traffic loss
Solution Approach 1:
The patent pre-calculates multiple optimal alternate paths before failure occurs, storing them for immediate use. This eliminates the need for complete recalculation when failures happen, maintaining route optimality through pre-computed paths while preventing data traffic loss by avoiding computation delays.
Solution Approach 2:
Instead of performing complete route recalculation, the patent uses pre-computed alternate paths that provide sufficient routing options without the excessive computation of full recalculation. This partial action approach maintains route optimality where needed while avoiding the productivity loss associated with complete recalculation.
4Reliability
If hierarchical routing topology with multiple paths is implemented, then resilience to link failure improves, but routing complexity increases
Solution Approach 1:
The patent segments the network into a hierarchical structure with parent and child supernodes, organizing multiple paths in a structured manner. This segmentation provides resilience through alternative paths while managing complexity by imposing a clear hierarchical framework on the routing topology.
Solution Approach 2:
The patent introduces a hierarchical dimension to the routing topology, organizing nodes into parent-child supernode relationships. This additional dimensional structure provides multiple paths for resilience while managing complexity through the hierarchical framework, making the system more organized and manageable despite the increased path options.
Data Source
Figure 1
Figure 2
Figure 3
AI summary
In one embodiment, a method comprises creating, in a computing network, a hierarchal routing topology for reaching a destination, the hierarchal routing topology comprising a single parent supernode providing reachability to the destination, and a plurality of child supernodes, each child supernode comprising one or more exit network devices each providing a corresponding link to the parent supernode; receiving, in one of the child supernodes, a data packet for delivery to the destination; causing the data packet to traverse along any available data link in the one child supernode independent of any routing topology established by network devices in the one child supernode, until the data packet reaches one of the exit network devices; and the one exit network device forwarding the data packet to the parent supernode, via the corresponding link, for delivery to the destination.