Network Link Optimization Profiles via Simulated Annealing

Resolve Bottlenecks,
Find Innovative SolutionsGenerate Solutions

Solution Overview

Problem

Existing network optimization profiles for VPN tunneling during roaming are often manually configured and lack specificity, leading to suboptimal performance due to limited expertise and inability to adapt to varying physical networks, resulting in potential degradation of network performance.

Innovation Solution

A user-guided method and system for generating network link optimization profiles that rank and test performance criteria, apply weighted configuration parameters using simulated annealing, and iteratively refine settings based on user preferences, ensuring optimal configuration for specific network conditions.

Engineering Contradictions & Design Principles

VSEngineering Contradiction Analysis

1Measurement precision

If network optimization profiles are manually created by vendors in a lab environment, then the profiles can be configured with expert knowledge, but the profiles are limited to the specific lab network conditions and cannot be effectively applied to different field networks

Engineering Contradiction:

Improveoptimization profile configuration accuracyVSAvoidnetwork condition adaptability

Core Design Contradiction:

Measurement precisionVSAdaptability or versatility

Solution Approach 1:

The system enables automatic generation of network optimization profiles through simulated annealing algorithms that self-adjust configuration parameters based on measured network performance, eliminating the need for manual vendor configuration and enabling adaptation to any network condition without requiring expert intervention

Inventive Principle:

Principle #25Self-service

Solution Approach 2:

The system dynamically changes multiple network configuration parameters simultaneously through simulated annealing optimization, allowing the profile to adapt to different network conditions by adjusting parameters such as TCP buffer sizes, packet timing, and protocol settings based on measured performance metrics

Inventive Principle:

Principle #35Parameter changes

2Adaptability or versatility

If end users manually tune optimization profiles to meet specific field needs, then the profiles can be customized for specific network conditions, but users lack the specialized expertise and may degrade network performance through misconfiguration

Engineering Contradiction:

Improvenetwork condition customizationVSAvoidnetwork performance stability

Core Design Contradiction:

Adaptability or versatilityVSReliability

Solution Approach 1:

The system measures actual network performance metrics and uses this feedback to automatically adjust optimization profile parameters through simulated annealing, ensuring that customization is based on objective performance data rather than user guesswork, thereby maintaining reliability while achieving adaptability

Inventive Principle:

Principle #23Feedback

Solution Approach 2:

The system replaces manual user tuning with an automated computational optimization algorithm (simulated annealing) that systematically explores parameter space to find optimal configurations, substituting human expertise with a reliable mathematical optimization process

Inventive Principle:

Principle #28Mechanics substitution (Replace mechanical system)

3Adaptability or versatility

If network optimization profiles are manually created in the field by operators, then the profiles can be adapted to actual network conditions, but the lack of expertise results in suboptimal configuration

Engineering Contradiction:

Improvefield network adaptabilityVSAvoidoptimization profile configuration quality

Core Design Contradiction:

Adaptability or versatilityVSManufacturing precision

Solution Approach 1:

The system performs self-configuration by automatically measuring network characteristics and generating optimized profiles through simulated annealing algorithms, eliminating the need for field operators to manually create profiles and ensuring high configuration quality without requiring expert knowledge

Inventive Principle:

Principle #25Self-service

Solution Approach 2:

The system performs preliminary network measurements and analysis before generating the optimization profile, gathering necessary network characteristic data in advance to enable accurate profile creation that adapts to field conditions while maintaining high configuration quality

Inventive Principle:

Principle #10Preliminary action

Data Source

PatentUS9112806B2User guided generation of network link optimization profiles

Publication Date: 2015.08.18 INTERNATIONAL BUSINESS MACHINE CORPORATION

AI summary

Embodiments of the present invention address deficiencies of the art in respect to optimization profile generation and provide a method, system and computer program product for user guided generation of network link optimization profiles. In one embodiment of the invention, a network optimization profile generation method can be provided. The method can include ranking different performance criterion for a target network, testing the target network for the different performance criterion, weighting results of the testing according to the ranking of the different performance criterion, generating a set of target network configuration parameters through optimization of the weighted results, for instance simulated annealing, and applying the set of target network configuration parameters to the target network as a profile.