Extended Remote LFA Backup Path Calculation
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing network routing technologies face challenges in achieving 100% failure coverage for single link and node failures, particularly in non-unit-cost networks, and are complex to implement, with existing methods requiring multiple addresses/labels and routing states, and not providing adequate support for shared risk link groups (SRLGs).
Innovation Solution
The extended remote Loop Free Alternates (LFA) process identifies intermediate nodes that do not traverse the failed node, using a unit-cost topology to provide 100% failure coverage for single link and node failures with only one extra address or label per node, and defines rules for tunnel endpoints that do not meet normal RLFA rules, enabling backup paths in arbitrary networks.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Reliability
If standard LFA or RLFA methods are used to compute backup paths, then failure coverage is improved, but device complexity and state requirements increase due to needing multiple addresses/labels per node
Solution Approach 1:
The patent makes a single backup address/label serve multiple purposes by enabling it to protect against multiple different failures (different next-hop nodes or links). Instead of requiring dedicated backup addresses for each failure scenario, one universal backup address can handle various failure types through the extended RLFA mechanism, thereby reducing the number of routing states needed while maintaining comprehensive failure coverage
Solution Approach 2:
The patent changes the cost parameters in the topology from actual link costs to unit costs (all links have equal cost of 1). This parameter transformation allows the backup path calculation to work correctly with a single backup address regardless of the actual varying link costs in the network, simplifying the routing state requirements while maintaining proper backup path selection
2Reliability
If standard LFA/RLFA rules are applied, then path validity is ensured, but coverage is insufficient for arbitrary networks particularly non-unit-cost networks
Solution Approach 1:
The patent transforms the network topology parameters by converting actual link costs to unit costs for backup path calculation purposes. This parameter change enables the extended RLFA mechanism to provide 100% failure coverage in arbitrary networks including non-unit-cost networks, as the unit cost assumption simplifies the mathematical conditions for backup path validity while the actual path selection still respects original cost metrics
Solution Approach 2:
The patent introduces an intermediate calculation layer using unit-cost topology that mediates between the actual varying-cost network and the backup path selection logic. This intermediate representation allows the system to apply simplified RLFA rules while still providing correct backup paths for the original arbitrary-cost network, thereby extending adaptability to all network topologies
3Measurement precision
If routing protocols recalculate routes after network failures, then routing accuracy is maintained, but convergence time increases causing traffic blocking or loss
Solution Approach 1:
The patent performs preliminary calculation of backup paths using extended RLFA with unit-cost topology before actual network failures occur. These backup paths are pre-computed and stored, allowing immediate switchover upon failure detection without waiting for routing protocol convergence. This preliminary action eliminates the time delay between failure and accurate routing restoration while maintaining routing precision through the mathematically guaranteed loop-free backup paths
4Productivity
If existing FRR technologies are implemented, then fast reroute capability is provided, but implementation complexity increases due to multiple addresses/labels requirements
Solution Approach 1:
The patent makes a single backup address/label universal by enabling it to protect against multiple different failures through extended RLFA. Instead of implementing complex mechanisms requiring multiple dedicated backup addresses for each failure scenario, the system uses one universal backup address that can be activated for various failure types, thereby maintaining fast reroute capability while dramatically reducing implementation complexity and configuration overhead
Data Source
Figure 1
Figure 2
Figure 3A
AI summary
A method is implemented by a network element or controller for determining a backup path for a fast reroute process to be utilized in response to a network event invalidating a primary path to a destination node. The method identifies at least one intermediate node that has a backup loop free alternative (LFA) path to a destination node in a network where no path meeting LFA conditions can be found for a point of local repair (PLR).