Network Controller Egress Planning With Staged Traffic Changes

Resolve Bottlenecks,
Find Innovative Solutions
Generate Solutions

Solution Overview

Problem

Manual selection of execution plans for network traffic assignment changes is impractical due to the large number of permutations and risks overloading network devices, making it difficult to achieve efficient and safe traffic reassignment.

Innovation Solution

A network controller automatically selects an execution plan using an optimization algorithm and heuristics to break down traffic assignment changes into manageable steps, minimizing costs and quality metrics to ensure safe and efficient network transitions.

Engineering Contradictions & Design Principles

VSEngineering Contradiction Analysis

1Reliability

If manual selection of execution plans is performed, then the network operator can control traffic assignment changes, but the process becomes impractical due to the large number of permutations and risks overloading network devices

Engineering Contradiction:
Improvesafe traffic reassignmentVSAvoidexecution plan selection
Core Design Contradiction:
ReliabilityVSEase of operation

Solution Approach 1:

The system enables automated selection of execution plans through optimization algorithms that independently evaluate multiple permutations and select the optimal plan without human intervention. The network controller automatically determines the best execution plan based on cost metrics and quality parameters, eliminating the impractical manual selection process while ensuring safe traffic reassignment.

Inventive Principle:
Principle #25Self-service

2Productivity

If automated execution plans are implemented, then the selection process becomes efficient, but network devices may be overloaded during transitions

Engineering Contradiction:
Improveexecution plan selection efficiencyVSAvoidnetwork device overloading
Core Design Contradiction:
ProductivityVSObject-affected harmful factors

Solution Approach 1:

The execution plan is divided into multiple intermediate steps that progressively transition traffic assignments from the first set to the second set. Each step modifies only a subset of traffic assignments, preventing sudden large-scale changes that could overload network devices. The optimization algorithm evaluates the entire sequence of steps to ensure smooth transitions while maintaining efficiency.

Inventive Principle:
Principle #1Segmentation

Solution Approach 2:

The system performs preliminary evaluation of multiple execution plans using optimization algorithms before implementation. Cost metrics and quality parameters are calculated in advance to identify the optimal execution plan that avoids overloading network devices. This preliminary analysis ensures that the selected plan efficiently transitions traffic assignments without causing harmful overloading effects.

Inventive Principle:
Principle #10Preliminary action

3Measurement precision

If multiple execution plans are evaluated, then the optimal plan can be selected based on cost and quality metrics, but the complexity of determining and comparing plans increases

Engineering Contradiction:
Improveexecution plan evaluation accuracyVSAvoidoptimization algorithm complexity
Core Design Contradiction:
Measurement precisionVSDevice complexity

Solution Approach 1:

The system uses optimization algorithms that evaluate execution plans based on configurable cost metrics and quality parameters. By changing and adjusting these parameters, the system can precisely measure and compare different execution plans to select the optimal one. The parameters allow flexible evaluation criteria while the automated algorithms manage the complexity of determining and comparing multiple plans.

Inventive Principle:
Principle #35Parameter changes

Data Source

PatentEP3855687B1Safely engineering egress traffic changes
Publication Date: 2025.09.10 JUNIPER NETWORKS INC
  • EP3855687B1 patent drawingFigure 1A
  • EP3855687B1 patent drawingFigure 1B
  • EP3855687B1 patent drawingFigure 2

AI summary

A network device can automatically select an execution plan from a set of possible execution plans that cause a first set of traffic assignments in a network to be changed to a second set of traffic assignments. A traffic assignment indicates assignments of the traffic to one or more tunnels, internal routes and/or peer links to be utilized for routing traffic received at provider edge routers through a network to prefixes. A traffic assignment can have various parameters such as bandwidth, transmission costs etc. Each execution plan has one or more steps, and each step has one or more traffic assignment changes progressing from the first set of traffic assignments to the second set of traffic assignments. The network device can automatically select an execution plan based on an evaluation metric determined for each execution plan. The evaluation metric can be a cost based metric or a quality based metric.