Rerouting Sequence Planning for LSP Adjustment

Resolve Bottlenecks,
Find Innovative Solutions
Generate Solutions

Solution Overview

Problem

Conventional rerouting sequence planning algorithms in distributed networks and software-defined networking (SDN) face challenges in optimizing network service deployment and maximizing network utilization, with existing methods experiencing poor time performance and low success ratios due to exponential time complexity and uneven network load distribution.

Innovation Solution

A rerouting sequence planning method that calculates reference values for label switched paths (LSPs) to determine adjustment priorities, allowing for the selection and adjustment of LSPs with the highest priority to either a final-state or temporary LSP, and iteratively processes remaining LSPs to improve time performance and success ratios.

Engineering Contradictions & Design Principles

VSEngineering Contradiction Analysis

1Reliability

If error trial and fallback algorithm is used to find adjustment sequence, then the adjustment can be completed, but the time performance deteriorates exponentially

Engineering Contradiction:
Improveadjustment successVSAvoidtime performance
Core Design Contradiction:
ReliabilityVSLoss of time

Solution Approach 1:

The patent calculates reference values for all LSPs in advance to determine their adjustment priorities before actually performing the adjustment operations. This preliminary calculation of priorities allows the system to execute adjustments in an optimized sequence without needing to retry failed adjustments, thereby avoiding exponential time complexity while maintaining high adjustment success rates.

Inventive Principle:
Principle #10Preliminary action

2Device complexity

If direct adjustment from initial network state to final network state is required, then the adjustment process is simplified, but the network utilization ratio requirement becomes excessively high

Engineering Contradiction:
Improveadjustment process complexityVSAvoidnetwork utilization ratio
Core Design Contradiction:
Device complexityVSQuantity of substance

Solution Approach 1:

The patent segments the adjustment process into multiple stages by introducing temporary LSPs as intermediate steps. Instead of requiring direct adjustment from initial to final state, the system allows LSPs to first transition to temporary states and then to final states, thereby reducing the network utilization ratio requirement at any single moment while maintaining overall adjustment effectiveness.

Inventive Principle:
Principle #1Segmentation

3Manufacturing precision

If GOA algorithm is used to calculate paths, then the path cost is minimized, but the path concentration on specific links increases and success ratio of rerouting decreases

Engineering Contradiction:
Improvepath cost optimizationVSAvoidrerouting success ratio
Core Design Contradiction:
Manufacturing precisionVSReliability

Solution Approach 1:

The patent introduces local quality differentiation by calculating reference values for each LSP based on its specific characteristics and the current network state. This allows the system to make localized adjustment decisions for each LSP rather than applying a uniform global optimization, thereby preventing path concentration on specific links while maintaining overall cost efficiency.

Inventive Principle:
Principle #3Local quality

Data Source

PatentEP3073686B1Rerouting sequence planning method and system
Publication Date: 2020.02.19 HUAWEI TECH CO LTD
  • EP3073686B1 patent drawingFigure 1
  • EP3073686B1 patent drawingFigure 2-a~2-b
  • EP3073686B1 patent drawingFigure 3-a~3-b

AI summary

Embodiments of the present invention disclose a rerouting sequence planning method and system, so as to improve time performance of a rerouting sequence planning algorithm and a success ratio of adjustment. The method in the embodiments of the present invention includes: calculating reference values of to-be-adjusted label switched paths LSPs, where the reference values are used to denote adjusted priorities of the to-be-adjusted LSPs; selecting a to-be-adjusted LSP with a highest priority from the to-be-adjusted LSPs; determining, according to a critical value of the reference values of the to-be-adjusted LSPs and a reference value of the to-be-adjusted LSP with the highest priority, whether the to-be-adjusted LSP with the highest priority is suitable for adjustment, where the critical value of the reference values of the to-be-adjusted LSPs denotes a minimum reference value required for a to-be-adjusted LSP to be adjusted to a final-state LSP; and if the to-be-adjusted LSP with the highest priority is suitable for adjustment, adjusting the to-be-adjusted LSP with the highest priority to a corresponding final-state LSP; and if the to-be-adjusted LSP with the highest priority is not suitable for adjustment, selecting at least one to-be-adjusted LSP and adjusting the at least one selected to-be-adjusted LSP to a corresponding temporary LSP.