MRT-FRR Endpoint Selection for Optimal Detour Paths
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Current IP Fast ReRoute (IPFRR) techniques, such as Loop-Free Alternates (LFA), struggle to provide full failure coverage in large networks due to their limitations in handling failure scenarios, leading to suboptimal detour paths and increased network complexity, and lack a clear solution for selecting optimal MRT endpoints.
Innovation Solution
A method and system for selecting an endpoint for a maximally redundant tree-fast reroute (MRT-FRR) detour path that calculates a clean set of unimpacted nodes and optimizes detour path cost or length by pre-calculating detour paths and configuring forwarding to a candidate node based on configured options, using a network processor with MPLS and IP modules to handle data packets.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Ease of manufacture
If Loop-Free Alternates (LFA) is used for IPFRR, then implementation simplicity is improved, but failure coverage is worsened
Solution Approach 1:
The patent segments the failure coverage problem into two parts: LFA handles local failures (simple implementation), while MRT-FRR handles remote failures (comprehensive coverage). This segmentation allows each technique to optimize for its specific scope, resolving the contradiction between simplicity and comprehensive failure coverage.
Solution Approach 2:
The patent merges LFA and MRT-FRR into a hybrid IPFRR system. LFA provides simple local failure protection, while MRT-FRR extends coverage to remote failures. The combination achieves comprehensive failure coverage while maintaining implementation feasibility through the use of existing LFA infrastructure.
2Reliability
If network topology is organized to become LFA friendly, then failure coverage is improved, but network complexity and expenditure are worsened
Solution Approach 1:
The patent introduces MRT-FRR as an intermediary technique that operates independently of LFA's topology requirements. Instead of redesigning the network topology to be LFA-friendly, MRT-FRR acts as a mediator that provides additional failure coverage using existing topological structures, thereby avoiding increased network complexity.
3Reliability
If MPLS FRR using RSVP is used, then fast reroute capability is improved, but scalability is worsened
Solution Approach 1:
The patent replaces the RSVP-based MPLS FRR mechanism with an IP-based MRT-FRR approach. Instead of using RSVP resource reservation and MPLS label switching, the system uses IP routing protocols and shortest path first algorithms to achieve fast reroute capability, thereby improving scalability in large networks.
4Length of moving object
If detour path cost is minimized, then path optimality is improved, but endpoint selection flexibility is worsened
Solution Approach 1:
The patent implements dynamic endpoint selection where the optimal endpoint is determined based on current network conditions and failure scenarios. The system can adaptively choose different endpoints from the clean set of nodes, allowing flexibility in endpoint selection while minimizing detour path cost through dynamic calculation and configuration.
Data Source
AI summary
A method is described to be implemented by a node in a network. The method is for selecting an endpoint for a maximally redundant tree-fast reroute (MRT-FRR) detour path to optimize detour path cost or length across the network. The method defines a set of steps including selecting a destination node and next hop failure to calculate detour paths. A clean set of nodes for the network is then calculated, where the clean set of nodes are nodes in the network that are not impacted in reaching the destination node by the failure in the given next hop. A candidate node for the endpoint of the detour p path is selected from the set of clean nodes based on any one of a plurality of configured options and forwarding of data packets is configured to the selected candidate as the endpoint of the detour path to the destination node.


