Loop-Free Alternate Next Hops for Multipoint LSP Fast Reroute

Resolve Bottlenecks,
Find Innovative Solutions
Generate Solutions

Solution Overview

Problem

Traditional link state protocols in computer networks take a long time to adapt to link or node failures, leading to delays in traffic rerouting and potential packet loss due to the need for routers to recalculate shortest paths.

Innovation Solution

Implementing loop-free alternate (LFA) next hops in multipoint label switched paths (LSPs) using point-to-point (P2P) LSPs with LFA, which allows for immediate rerouting of multicast traffic through a label stack including both P2P and multipoint LSP labels, enabling fast reroute (FRR) until network convergence.

Engineering Contradictions & Design Principles

VSEngineering Contradiction Analysis

1Reliability

If traditional link state protocols are used for routing, then routing information is maintained accurately, but failure recovery time is long due to shortest path recalculation

Engineering Contradiction:
Improverouting accuracyVSAvoidfailure recovery time
Core Design Contradiction:
ReliabilityVSLoss of time

Solution Approach 1:

The patent pre-calculates and installs alternate next-hop routes before failures occur. When a link or node failure is detected, the router immediately switches to the pre-computed alternate route without waiting for shortest path recalculation, thus resolving the contradiction between maintaining routing accuracy and reducing failure recovery time

Inventive Principle:
Principle #10Preliminary action

Solution Approach 2:

The patent introduces an intermediary mechanism (alternate next-hop cache) that stores pre-computed backup routes. This intermediary allows the routing system to maintain accuracy through traditional link-state protocols while providing fast failover through pre-prepared alternate paths, eliminating the need for immediate shortest path recalculation upon failure

Inventive Principle:
Principle #24Intermediary (Mediator)

2Reliability

If shortest path algorithm is re-computed upon failure, then accurate routing is maintained, but traffic forwarding is delayed during convergence

Engineering Contradiction:
Improverouting accuracyVSAvoidtraffic forwarding speed
Core Design Contradiction:
ReliabilityVSProductivity

Solution Approach 1:

The system performs preliminary computation of alternate routes and installs them in the forwarding engine before failures occur. This allows immediate traffic forwarding through pre-computed paths when failures happen, maintaining routing accuracy while eliminating convergence delays for traffic forwarding

Inventive Principle:
Principle #10Preliminary action

Solution Approach 2:

The patent creates copies of routing information in the form of pre-computed alternate next-hop routes stored in the forwarding engine. These copies enable fast traffic forwarding without requiring real-time shortest path recalculation, thus maintaining routing accuracy while improving traffic forwarding speed during failures

Inventive Principle:
Principle #26Copying

3Loss of time

If alternate next-hop is pre-selected, then failure recovery is faster, but additional routing complexity is introduced

Engineering Contradiction:
Improvefailure recovery timeVSAvoidrouting complexity
Core Design Contradiction:
Loss of timeVSDevice complexity

Solution Approach 1:

The patent introduces an intermediary structure (alternate next-hop cache) that simplifies the complexity management. The cache stores pre-computed backup routes in a structured manner, allowing fast failure recovery while managing routing complexity through organized storage and retrieval mechanisms rather than complex real-time calculations

Inventive Principle:
Principle #24Intermediary (Mediator)

4Reliability

If traditional convergence process is used, then routing tables are updated accurately, but packet loss occurs during the update period

Engineering Contradiction:
Improverouting table accuracyVSAvoidpacket loss
Core Design Contradiction:
ReliabilityVSLoss of substance

Solution Approach 1:

The patent performs preliminary installation of alternate next-hop routes in the forwarding engine before failures occur. When failures happen, traffic is immediately redirected to pre-prepared alternate paths, eliminating packet loss during the convergence period while maintaining routing table accuracy through coordinated updates

Inventive Principle:
Principle #10Preliminary action

Solution Approach 2:

The system provides beforehand cushioning by pre-computing and storing alternate routes that act as a buffer against failures. This cushioning mechanism ensures continuous traffic forwarding through alternate paths during failures, preventing packet loss while the routing tables are being updated to maintain accuracy

Inventive Principle:
Principle #11Beforehand cushioning (Prior cushioning)

Data Source

PatentEP2663040B1Fast reroute using loop free alternate next hops for multipoint label switched paths
Publication Date: 2017.07.12 JUNIPER NETWORKS INC
  • EP2663040B1 patent drawingFigure 1
  • EP2663040B1 patent drawingFigure 2
  • EP2663040B1 patent drawingFigure 3

AI summary

The techniques of this disclosure provide local protection for network traffic in multipoint label switched paths (LSPs) due to link or node failure using loop-free alternate (LFA) next hops. The techniques include establishing a vanilla or point-to-point (P2P) LSP with LFA next hops between routers of a multipoint LSP for use in the event of link or node failure in the multipoint LSP. Upon a failure, the multicast traffic is tunneled between the routers using the P2P LSP with LFA to an alternate next hop with an associated label stack. The techniques of this disclosure define the label stack as including a P2P LSP label as well as a multipoint LSP label. In this way, the P2P LSP with LFA may be used for fast reroute (FRR) of traffic in the multipoint LSP until a convergence process completes for a new multipoint branch of the multipoint LSP.