Network Resource Orchestration via Predictive ML Adaptation
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Current network resource management approaches are reactive and lack a proactive strategy for scaling network capacity, and are limited by silos between data sources, making it difficult to implement optimal network service delivery in dynamic environments.
Innovation Solution
A method that determines a desired performance level for network services, generates predictions based on internal and external data, identifies decision scenarios for resource orchestration, and adapts resources accordingly, while also providing a system for proactive maintenance planning using machine-learning algorithms.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Reliability
If reactive approach to increase and/or decrease in network resource capacity deployment is used, then network resources can be adjusted based on actual performance triggers, but network service delivery cannot be optimized proactively in dynamic environments
Solution Approach 1:
The system performs preliminary actions by proactively predicting future network service patterns using machine learning algorithms before actual demand occurs. The orchestrator proactively adjusts network resource capacity based on predicted patterns, rather than waiting for reactive triggers, enabling optimization in dynamic environments ahead of time
Solution Approach 2:
The system implements feedback mechanisms by continuously monitoring actual network performance and comparing it with predicted patterns. This feedback loop allows the orchestrator to refine predictions and adjust resource allocation dynamically, improving both reliability and adaptability through continuous learning and adjustment
2Adaptability or versatility
If silos between various data sources are maintained, then data security and system simplicity are preserved, but ability to implement proactive network resource capacity deployment is limited
Solution Approach 1:
The system introduces an intermediary component - the machine learning-based prediction module - that sits between various data sources and the network orchestrator. This intermediary aggregates and processes data from multiple sources without requiring direct integration between them, enabling proactive resource deployment while managing architectural complexity through a centralized mediation layer
3Productivity
If manual interventions are required to upscale or downscale network resources, then network function deployment can be controlled precisely, but fast deployment of new network functions is prevented
Solution Approach 1:
The system implements self-service by enabling network resources to automatically scale up or down based on predicted service patterns. The orchestrator autonomously makes deployment decisions without requiring manual interventions, while maintaining precise control through algorithmic prediction models that account for network requirements and constraints
Data Source
AI summary
Methods and systems for managing network resources enabling network services over a network and for managing maintenance of network resources. The method comprises determining a desired performance level for the network services, the desired performance level being associated with service metrics that establish compliance with a service level agreement; accessing internal data relating to operations of the network and external data not relating to operations of the network; generating a prediction of a network service pattern based on the desired performance level of the network services, the internal data and the external data; identifying a decision scenario for orchestration of the network resources, the decision scenario establishing a configuration of the network resources, the decision scenario being generated based on a correlation of the prediction of the network service pattern and availability of the network resources; and causing to adapt the network resources based on the decision scenario.