Network Topology Discovery Using Time-Series Graph Mapping
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Current computer network topology mapping tools are unable to automatically discover, store, and provide quick access to network object topologies at a large scale, often requiring sensitive information access, failing to perform repeated discovery, and not providing a comprehensive graphical representation with time dimension.
Innovation Solution
A system and method for continual automated discovery of network topology, utilizing a collector, scheduler, and processors to collect and process configuration and time-series information from network objects, creating a graph representation of the topology over time.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Extent of automation
If current topology mapping tools are used, then basic topology mapping is possible, but they cannot automatically discover and store topology at large scale
Solution Approach 1:
The system segments the topology discovery process into multiple components: a collector that gathers data from network objects, a scheduler that manages discovery operations, and processors that analyze and store topology information. This segmentation enables automated large-scale topology mapping by distributing functions across specialized modules, each handling specific aspects of the discovery process independently and efficiently.
Solution Approach 2:
The patent introduces a collector as an intermediary component that sits between network objects and the processing system. This collector automatically gathers topology data from network objects without requiring direct access to sensitive information or manual intervention, enabling automated discovery while filtering and preparing data for further processing and storage.
2Reliability
If repeated discovery is performed to track changes, then topology accuracy improves, but system resource consumption increases
Solution Approach 1:
The system implements periodic topology discovery operations scheduled at appropriate intervals rather than continuous monitoring. The scheduler manages these periodic actions to detect changes while optimizing resource consumption, performing discovery only when necessary to maintain accuracy and track topology evolution over time without exhausting system resources.
Solution Approach 2:
The system uses feedback mechanisms where the processor analyzes discovered topology data, compares it with stored information, and determines whether changes have occurred. This feedback loop enables the system to adjust its discovery frequency and intensity based on actual topology stability, reducing unnecessary resource consumption while maintaining reliable change detection when needed.
3Loss of information
If comprehensive data collection is performed, then topology completeness improves, but data processing complexity increases
Solution Approach 1:
The data processing function is segmented into specialized processors that handle different aspects of topology data independently. Each processor focuses on specific tasks such as data validation, relationship mapping, or storage optimization, reducing the complexity burden on any single component while maintaining comprehensive data collection and processing capabilities.
Data Source
AI summary
There are disclosed devices, system and methods for continual automated discovering of a topology of a computer network of physical network objects for an application. A first physical network object selects portions of sets of data messages being sent over time by the application and related network objects; and collects network configuration and time dimension information, and timeseries information from the portions of messages. A second physical network object receives the information and uses it to determine topology information for the application over time, which includes unique keys of, types of objects of, types of relationships between pairs of, groupings of, time dimension data of and metrics data of the physical network objects. The topology information can be stored; and can be queried to create and display a graph representation of the topology information that changes over time.