Head-End Node Failure Protection in TE-LSP Mesh Networks
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Current methods lack effective protection mechanisms for failures of the head-end node in TE-LSPs, particularly in P-to-P TE-LSP mesh networks, which can lead to prolonged network downtime and scalability issues.
Innovation Solution
A neighboring upstream node, or protecting node, learns about primary TE-LSPs and establishes repair TE-LSPs to bypass the head-end node failure by rerouting traffic through repair TE-LSPs and refreshing state control blocks, allowing for quick recovery without manual configuration.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Reliability
If conventional protection methods are used for TE-LSPs, then protection against link failures is provided, but protection against head-end node failures is not available
Solution Approach 1:
The patent implements preliminary action by having the protecting node pre-establish repair TE-LSPs to next-next-hops before the head-end node failure occurs. The protecting node learns about primary TE-LSPs and establishes backup paths in advance, so when failure happens, traffic can be immediately rerouted without waiting for failure detection and path computation.
Solution Approach 2:
The patent introduces an intermediary mechanism where the protecting node acts as a mediator between the failed head-end node and the rest of the network. The protecting node receives state control blocks from the head-end node and replicates them to maintain TE-LSP states without the original head-end node, enabling continuous operation through the intermediary protecting node.
2Reliability
If manual configuration is used for protection, then protection can be set up, but the configuration process is cumbersome and time-consuming
Solution Approach 1:
The patent implements self-service by enabling the network to automatically discover and configure protection paths without manual intervention. The protecting node automatically learns about primary TE-LSPs through IGP advertisements or fake TE-LSPs, automatically establishes repair TE-LSPs, and automatically activates them when failure occurs, eliminating the need for manual protection configuration.
Solution Approach 2:
The patent uses feedback mechanisms where the head-end node sends state control blocks (such as RSVP Path messages) to the protecting node, which then uses this feedback information to automatically establish repair TE-LSPs. The system continuously monitors TE-LSP states and automatically adjusts protection paths based on real-time network conditions.
3Reliability
If PE-to-PE TE-LSP mesh networks are deployed, then protection coverage is extended, but the number of TE-LSPs increases and scalability is reduced
Solution Approach 1:
The patent segments the protection function by introducing protecting nodes at strategic locations (P routers) rather than requiring full mesh at all nodes. Each protecting node independently manages repair TE-LSPs for specific primary TE-LSPs, dividing the overall protection management into smaller, manageable segments that reduce the total number of TE-LSPs while maintaining comprehensive protection coverage.
Solution Approach 2:
The patent makes protecting nodes multi-functional by enabling them to perform multiple roles: they act as regular routers for normal traffic, serve as protection points for head-end node failure, and manually trigger repair paths when needed. This universality reduces the need for dedicated protection infrastructure and minimizes the number of TE-LSPs required.
4Reliability
If repair TE-LSPs are established to next-next-hops, then recovery scope is enhanced, but state control block management becomes more complex
Solution Approach 1:
The patent uses copying by having the protecting node replicate state control blocks from the failed head-end node to maintain TE-LSP states. Instead of managing complex state synchronization, the protecting node simply copies the state information (such as RSVP Path messages with next-next-hop addresses) to establish and maintain repair TE-LSPs, simplifying state management while enhancing recovery scope.
Data Source
AI summary
In one embodiment, a repair label switched path (LSP) is established for a primary LSP having a head-end node. The repair LSP extends from a neighboring upstream node of the head-end node to a downstream neighboring node of the head-end node. When a failure of the head-end node is detected, the neighboring upstream node reroutes traffic onto the repair LSP. The rerouted traffic rejoins the primary LSP at the downstream neighboring node. The neighboring upstream node refreshes state of the primary LSP to maintain the primary LSP after failure of the head-end node.