Predictive Routing Efficacy Estimation via Metadata Analysis
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Network operators are hesitant to adopt predictive routing due to concerns about its efficacy, as existing methods require extensive data enrollment and ingestion for thousands of deployments, making them time-ineffective and non-scalable.
Innovation Solution
A device estimates performance metrics for predictive routing by obtaining metadata from existing network deployments, identifying network topologies not using predictive routing, and sending report data to a user interface, allowing for the evaluation of potential benefits and savings.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Reliability
If predictive routing is deployed in a prospective network, then routing efficacy and performance improvement are achieved, but extensive data enrollment and ingestion time is required
Solution Approach 1:
The system performs preliminary actions by pre-processing and storing routing metadata from multiple networks in advance. When evaluating a prospective network, the pre-processed metadata is quickly matched against the network topology to estimate predictive routing efficacy, eliminating the need for time-consuming data enrollment and ingestion for each new deployment evaluation.
2Measurement precision
If predictive routing is evaluated for each prospective network individually, then accurate efficacy estimation is achieved, but the process does not scale to thousands of deployments
Solution Approach 1:
The system creates a universal metadata repository that stores routing information from multiple different networks in a standardized format. This universal structure allows the same evaluation process to be applied across thousands of prospective networks simultaneously, enabling both accurate efficacy estimation and high scalability without requiring individualized processing for each deployment.
3Measurement precision
If telemetry data is collected for sufficient time for the predictive routing model to learn patterns, then model accuracy is improved, but the time commitment exceeds operator willingness
Solution Approach 1:
Instead of collecting and learning from telemetry data in the prospective network itself, the system creates copies of routing metadata from multiple existing networks that have already been processed by predictive routing models. These copied metadata patterns are then used to quickly estimate efficacy for the prospective network, achieving model accuracy without requiring a lengthy local learning period.
Data Source
AI summary
In one embodiment, a device obtains metadata for routing decisions made by a predictive routing service for a plurality of network deployments. The device identifies a network topology for a network deployment that does not use the predictive routing service. The device estimates, based on the metadata for routing decisions made by the predictive routing service, performance metrics for the predictive routing service were it to be used to make routing decisions for the network topology. The device sends, to a user interface, report data indicative of the performance metrics estimated for the predictive routing service were it to be used to make routing decisions for the network topology.