Network Assurance Service Predicting Configuration Impact
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing network assurance systems face challenges in detecting and analyzing performance degradation caused by configuration changes in networking devices, particularly due to the complexity and vast number of possible configuration scenarios, which can lead to catastrophic failures and delayed issue detection.
Innovation Solution
A machine learning-based approach that monitors network configuration changes, trains models on performance indicators, predicts the impact of configuration changes, and automatically applies or recommends changes to mitigate performance degradation, using components like a change detector engine, change impact predictor, and degradation analyzer.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Reliability
If machine learning models are used to predict configuration change impacts, then network reliability is improved, but device complexity increases
Solution Approach 1:
A machine learning model serves as an intermediary between configuration changes and network performance outcomes. The model is trained on historical configuration data and performance metrics to predict the impact of proposed changes before deployment, allowing network operators to assess risks without directly implementing potentially harmful configurations.
Solution Approach 2:
The system performs preliminary analysis of configuration changes by training machine learning models on historical data and using these models to predict outcomes before actual deployment. This advance prediction capability allows operators to prevent problematic changes before they affect network reliability.
2Measurement precision
If comprehensive network monitoring is implemented, then measurement precision is improved, but loss of time increases
Solution Approach 1:
The system collects and processes network performance data in advance, training machine learning models on historical configurations and outcomes. This preliminary processing creates ready-to-use predictive models that can quickly assess new configuration changes without requiring time-consuming analysis at the moment of deployment.
Solution Approach 2:
The system creates simplified representations of complex network states through machine learning models. Instead of analyzing raw network data directly when evaluating configuration changes, the system uses trained models that capture essential patterns, enabling faster decision-making while maintaining measurement precision.
Data Source
AI summary
In one embodiment, a network assurance service that monitors one or more networks receives data indicative of networking device configuration changes in the one or more networks. The service also receives one or more performance indicators for the one or more networks. The service trains a machine learning model based on the received data indicative of the networking device configuration changes and on the received one or more performance indicators for the one or more networks. The service predicts, using the machine learning model, a change in the one or more performance indicators that would result from a particular networking device configuration change. The service causes the particular networking device configuration change to be made in the network based on the predicted one or more performance indicators.