Local ML Model Switching for Network Assurance Constraints
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing network assurance systems face challenges in deploying machine learning models locally due to resource constraints and the need for accurate model performance without transmitting telemetry data externally, leading to suboptimal model execution and resource consumption.
Innovation Solution
A system for dynamically selecting and deploying machine learning models locally within a network assurance service, based on performance requirements and input features, to optimize model quality and resource usage, with the option to trigger custom model computation in the cloud when necessary.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If a machine learning model is deployed locally to assess network measurements, then model performance and accuracy are improved, but resource consumption increases
Solution Approach 1:
The system dynamically changes model parameters including selecting different models from a library based on current resource availability and performance requirements. The controller monitors resource consumption and adjusts which machine learning model is deployed locally, switching between models with different computational complexities to optimize the tradeoff between performance and resource usage.
2Adaptability or versatility
If multiple machine learning models are maintained for different network conditions, then adaptability is improved, but device complexity increases
Solution Approach 1:
The system segments the machine learning model library into multiple discrete models, each optimized for specific network conditions or assessment tasks. The controller selects and deploys only the relevant model segment needed for current network conditions, rather than maintaining all models simultaneously, reducing device complexity while preserving adaptability.
Solution Approach 2:
The machine learning model library serves multiple functions by containing models that can handle different network assessment scenarios. A single library structure provides universal access to various models, allowing the system to adapt to different conditions without requiring separate dedicated systems for each scenario.
3Reliability
If model performance is monitored and models are replaced when performance degrades, then reliability is improved, but loss of time occurs during model replacement
Solution Approach 1:
Multiple machine learning models are pre-trained and stored in the model library before deployment. When performance degradation is detected, the controller can immediately switch to a pre-prepared alternative model from the library without needing to train or download a new model, significantly reducing replacement time while maintaining reliability.
Data Source
AI summary
In one embodiment, a network assurance service uses a first machine-learning based model that is locally deployed to a network to assess a set of input features comprising measurements from the network. The service monitors, locally in the network, performance of the first machine learning-based model. The service determines that the monitored performance of the first machine learning-based model does not meet one or more performance requirements associated with the network. The service selects a second machine learning-based model for deployment to the network, based on the one or more performance requirements associated with the network and on the set of input features of the first machine learning-based model. The service deploys the selected second machine learning-based model to the network as a replacement for the first machine learning-based model.