Network Topology Graph from Telemetry Data
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing network topology determination and visualization systems are limited to site-level views, require substantial user input, and cannot provide comprehensive or historical views of network topologies, especially in distributed networks with changing connectivity.
Innovation Solution
A network analysis system that automatically generates a temporal network topology graph from telemetry data received from multiple network devices, representing both current and past connectivity and properties, without requiring substantial user input, by storing entity and connectivity information as nodes and edges in a graph database.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Device complexity
If existing systems provide site-level network topology views, then the complexity of managing distributed network devices is reduced, but the comprehensiveness of network topology information is limited
Solution Approach 1:
The patent merges multiple site-level topology views into a single organization-level topology graph that encompasses all network devices across all sites. The system combines distributed telemetry data from numerous devices into a unified comprehensive view, allowing users to see both individual site details and the overall organizational network structure simultaneously.
2Measurement precision
If existing systems require substantial user input to determine network topology, then the accuracy of topology information can be improved, but the ease of operation deteriorates
Solution Approach 1:
The system automatically collects telemetry data from network devices and uses machine learning models to determine topology relationships without requiring user input. The network devices self-report their connectivity information, and the system processes this data automatically to build and update the organization-level topology graph, eliminating manual configuration requirements.
3Loss of information
If existing systems provide current network topology views, then the relevance of topology information is improved, but the ability to perform historical analysis deteriorates
Solution Approach 1:
The system continuously collects and stores telemetry data from network devices over time, building a historical record of network topology changes. This preliminary data collection enables both current topology visualization and historical analysis, allowing users to examine how the network has evolved and perform root cause analysis on past issues.
Data Source
AI summary
An example network analysis system includes a memory storing telemetry data received from a plurality of network devices, the plurality of network devices includes extract entity information and connectivity information from the received telemetry data, wherein the entity information represents one or more network devices of the plurality of network devices and the connectivity information represents network connections between one or more devices of the plurality of network devices; and store the connectivity information and entity information as a network topology graph in a graph database, wherein the entity information is stored as nodes of the network topology graph and the connectivity information is stored as edges of network topology graph, and wherein the network topology graph represents an organization level topology of the organization network.