LDP Multicast Fast Reroute Loop Prevention
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Multicast networks face issues with transient micro loops and fast reroute loops during network changes, leading to potential network overload and data loss, especially in IP networks where control and data plane inconsistencies cause traffic to be sent back and forth between nodes using old and new topologies.
Innovation Solution
Implementing a method that involves ordered installation of the multicast Forwarding Information Base (mFIB) and Reverse Path Forwarding (RPF) checks, along with allocation of new labels for each ingress change, to ensure traffic is forwarded only when all nodes are converged, and using disjoint trees to prevent looping between old and new paths.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Speed
If fast reroute is implemented to quickly switch to backup paths during network changes, then network recovery speed is improved, but transient micro loops and fast reroute loops can occur causing traffic loss and network overload
Solution Approach 1:
The patent applies preliminary action by pre-distributing backup labels and establishing backup paths before network failures occur. When a failure is detected, the fast reroute mechanism can immediately switch to the pre-established backup path using the pre-distributed labels, eliminating the need for reactive path computation and enabling sub-50ms recovery while avoiding transient loops through coordinated label switching across the network
Solution Approach 2:
The patent implements feedback mechanisms through loop detection and prevention protocols that monitor traffic flow during fast reroute operations. When potential loops are detected, the system provides feedback to adjust label distribution and path selection, ensuring that backup paths do not create transient micro loops or fast reroute loops that would cause traffic loss or network overload
2Adaptability or versatility
If nodes switch to new topology paths during network changes, then network adaptability is improved, but control and data plane inconsistencies cause traffic to loop between old and new paths
Solution Approach 1:
The patent applies preliminary action by pre-distributing backup labels and establishing backup paths before network failures occur. When a failure is detected, the fast reroute mechanism can immediately switch to the pre-established backup path using the pre-distributed labels, eliminating the need for reactive path computation and enabling sub-50ms recovery while avoiding transient loops through coordinated label switching across the network
Solution Approach 2:
The patent uses label distribution protocol (LDP) messages and path vectors as intermediaries to coordinate the switch from old to new paths across all nodes. These intermediary mechanisms ensure that all nodes converge to the new topology simultaneously, preventing control and data plane inconsistencies that would otherwise cause traffic looping between old and new paths
Data Source
AI summary
A computer apparatus comprising a processor and a forwarding engine arranged to forward LDP multicast traffic along a multicast tree having a primary and a backup path in a converged network topology, the processor being configured to cause the forwarding engine to forward traffic via the backup path upon a topology change and send a changed topology label and path vector to at least one neighbor node in the changed topology.