Network Topology Mapping via Macro-Cluster Segmentation
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Current computer network topology mapping tools are unable to automatically discover and map relationships between macro-clusters at a large scale, often requiring sensitive information access, failing to perform repeated discoveries, and lacking the ability to group resources or collect critical metrics, and they do not provide a graphical representation that includes time dimension data or maps relationships between macro-clusters.
Innovation Solution
A system and method for continual automated discovery of network topology, which includes a discovery engine that accesses network objects to map relationships between macro-clusters, using a collector, scheduler, and data sources to gather configuration, metrics, and flow logs, processing this data to create a graphical representation of topology changes 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 map relationships between macro-clusters at a large scale
Solution Approach 1:
The system segments the network topology into macro-clusters, which are groups of network objects (such as servers, databases, network devices) that are related through common communication patterns or functional dependencies. This segmentation enables the system to handle large-scale networks by organizing them into manageable units that can be automatically discovered and mapped without requiring manual intervention or sensitive information access.
2Ease of operation
If existing tools are used, then topology mapping can be performed, but they require access to sensitive information
Solution Approach 1:
The system performs self-service by automatically discovering and mapping topology relationships without requiring explicit permission or access to sensitive information from network objects. The discovery engine analyzes network traffic patterns, communication metadata, and operational parameters to infer topology structures, eliminating the need for sensitive information access while maintaining accurate topology mapping capabilities.
3Productivity
If existing tools are used, then single-point topology mapping is possible, but they do not perform repeated discovery or compare changes over time
Solution Approach 1:
The system implements continuous topology discovery and mapping through repeated execution of the discovery engine over time. This continuous action enables the system to detect changes in network topology, identify evolving relationships between macro-clusters, and maintain an updated topological model without requiring manual re-mapping operations, thereby eliminating time loss associated with repeated manual analysis.
4Quantity of substance
If existing tools are used, then basic resource collection is possible, but they do not automatically group resources or collect critical metrics for each network object
Solution Approach 1:
The system merges multiple functions into a unified discovery engine that simultaneously collects resource information, groups network objects into macro-clusters, and aggregates critical metrics. This consolidation of functions into a single integrated system enables automatic resource grouping and metric collection without increasing overall system complexity, as the discovery engine handles all tasks through coordinated analysis of network data.
5Loss of information
If existing tools are used, then basic topology data can be obtained, but they do not provide a graph representation that mixes data with time dimension
Solution Approach 1:
The system adds a time dimension to the topology graph representation, transforming static topology data into a dynamic temporal graph that captures how network relationships evolve over time. This dimensional enhancement allows the system to represent both spatial topology and temporal changes simultaneously, providing comprehensive topology information without requiring separate analysis tools or increasing overall system complexity.
Data Source
AI summary
There are disclosed devices, system and methods for mapping relationships between macro-clusters of a network object topology of a computer communication network. A remote network object of the network is selected that has a relationship with one macro-cluster that has a relationship with another macro-cluster. Flow log data, metric data and configuration data are gathered from at least the selected network object. Configuration data and time data are generated for the sets network objects of the two macro-clusters using the gathered flow log data, metric data and configuration data. Network topology information is created using the configuration data and time data. The network topology information includes topology information for the relationship between the macro-clusters, for each macro-cluster and for the sets of network objects of the macro-clusters. The topology information can be stored and used to determine whether performance issues occur in the macro-clusters or relationship over time.