Network Flow Controller Using Custom Filter-Based Packet Sampling
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing network solutions lack granular analysis and control of network data, leading to overwhelming packet loads and unnecessary flow information, as they do not effectively identify the start or end of a flow and send the entire packet stream with minimal flexibility.
Innovation Solution
The implementation of a network management system with network flow controllers that dynamically track and control network traffic by identifying defined characteristics, reducing load on network devices, and improving security and telemetry processing through dynamic filter adjustments and flow group management.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If existing network solutions send entire packet streams with minimal flexibility, then flow information is captured, but processing load becomes overwhelming and granular analysis is lost
Solution Approach 1:
The patent extracts and exports flow information separately from the entire packet stream. Flow controllers identify flows and extract their characteristics (source/destination IPs, ports, protocols) to create flow records that are exported to telemetry systems, while the actual packet streams continue flowing through the network without being fully captured for analysis.
Solution Approach 2:
The patent segments the network traffic into identifiable flows using flow classification rules. Each flow is tracked independently with its own flow record containing specific characteristics, allowing selective monitoring and analysis of individual flows rather than processing the entire packet stream as a monolithic data set.
2Loss of information
If existing network solutions send entire packet streams, then complete flow data is available, but granular control and flexibility are reduced
Solution Approach 1:
The patent implements dynamic flow control where flow classification rules, sampling rates, and export policies can be adjusted in real-time based on network conditions and analysis requirements. The system can dynamically add, modify, or remove flow tracking rules without disrupting overall network operations, providing flexible control over what flow information is captured and exported.
Solution Approach 2:
The patent allows modification of flow characteristics and classification parameters through configurable rules. Flow controllers can adjust matching criteria, sampling intervals, and export thresholds to optimize between data completeness and processing efficiency, enabling granular control over the monitoring process.
3Quantity of substance
If flow telemetry processing systems receive excessive packet loads, then comprehensive coverage is achieved, but system performance and response time deteriorate
Solution Approach 1:
The patent applies partial action by selectively monitoring and exporting only specific flow characteristics rather than capturing entire packet streams. Flow controllers sample flow information at controlled rates and export only relevant flow records to telemetry systems, providing sufficient coverage for analysis while maintaining manageable data volumes and fast processing response times.
Data Source
AI summary
Methods, systems, and apparatuses for efficient tracking and exporting of network flows using custom filters. A network device receives network data and classifies a flow of the network data as belonging to a defined flow group from among a plurality of defined flow groups. A set of rules is identified that correspond to the defined flow group. Flow information is updated for a characteristic of the flow according to the set of rules. The network device determines, based on the flow information, that the flow satisfies a set of criteria involving the characteristic and, as a result of the determination, controls transmission of data associated with the flow.