Network Service Runtime Reconfiguration for Availability Constraints
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing network service deployment methods fail to guarantee continued compliance with availability and disruption requirements due to changes in infrastructure resources over time, such as failures, aging, or load redistribution, after initial deployment.
Innovation Solution
A method utilizing machine learning and mathematical models constructed at design-time to simulate infrastructure changes, generating configuration parameters that are used to train a runtime adaptation module to adjust network service settings dynamically, ensuring compliance with availability and disruption requirements.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Reliability
If design-time method is used to determine deployment options, then availability and disruption requirements are guaranteed initially, but the system cannot adapt to infrastructure changes over time
Solution Approach 1:
The patent transforms the static design-time configuration into a dynamic system that continuously monitors infrastructure resource characteristics and automatically adjusts configuration parameters at runtime. The system transitions from fixed deployment options to adaptive configuration that responds to changing infrastructure conditions, resolving the contradiction between initial reliability guarantee and runtime adaptability.
Solution Approach 2:
The patent implements a feedback mechanism where the system monitors actual infrastructure resource characteristics and compares them against the deployment options selected at design-time. When deviations are detected, the system triggers re-evaluation and adjustment of configuration parameters, creating a closed-loop control system that maintains availability guarantees despite infrastructure changes.
2Adaptability or versatility
If infrastructure resources change over time due to failures, aging, or load redistribution, then system adaptability improves, but compliance with availability constraints deteriorates
Solution Approach 1:
The patent performs preliminary actions by pre-calculating multiple deployment options at design-time with associated configuration parameters that would satisfy availability constraints under different infrastructure conditions. When runtime changes occur, the system can quickly switch to pre-computed alternative configurations rather than performing complex real-time optimization, thus maintaining availability compliance while adapting to infrastructure changes.
Solution Approach 2:
The patent changes configuration parameters dynamically based on monitored infrastructure resource characteristics. Instead of changing the fundamental system architecture or deployment topology, the system adjusts configurable parameters within existing deployment options to maintain availability compliance, enabling adaptation through parameter tuning rather than structural reconfiguration.
3Reliability
If configuration parameters are adjusted dynamically at runtime, then service availability is maintained under changing conditions, but system complexity increases
Solution Approach 1:
The patent creates simplified copies or representations of the complex infrastructure state through monitoring modules that track resource characteristics. Rather than directly managing complex infrastructure changes, the system works with simplified models and metrics that capture essential state information, enabling runtime adaptation without proportionally increasing system complexity.
Solution Approach 2:
The patent segments the runtime adaptation system into distinct functional modules: monitoring modules that detect infrastructure changes, evaluation modules that assess impact on availability constraints, and adjustment modules that modify configuration parameters. This segmentation allows each component to remain relatively simple while the integrated system provides comprehensive runtime adaptation capability.
Data Source
AI summary
There is provided a method of runtime adaptation of a network service (NS), a corresponding network node and non-transitory computer readable media. The method comprises detecting a deviation from an availability constraint imposed on resources for the NS. The method comprises executing a runtime adjustment model to determine new values for adjustable configuration parameters of the NS. The method comprises reconfiguring the adjustable configuration parameters of the NS according to the new values.


