SD-WAN Synthetic Traffic Recognition for Underlay Path Enrichment
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing network applications fail to provide comprehensive end-to-end metrics for SD-WAN networks due to the abstraction of the underlay network, limiting the visibility and monitoring capabilities of network performance.
Innovation Solution
The proposed technology integrates data from network monitoring applications with SD-WAN networks by identifying synthetic flows outside the SD-WAN and collecting data on these flows as they traverse the SD-WAN network, using a path insight tool to combine SD-WAN edge routers and underlay hops with network metrics, providing a comprehensive view of network paths.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Device complexity
If SD-WAN networks use abstraction of underlay network to simplify management, then device complexity is reduced, but measurement precision of end-to-end network metrics deteriorates
Solution Approach 1:
The patent introduces an intermediary system consisting of probes deployed at network edges and a centralized collector that mediates between the abstracted SD-WAN underlay and the need for detailed end-to-end metrics. The probes collect metrics from synthetic traffic flows and transmit them to the collector, which aggregates and correlates data to provide comprehensive visibility without requiring direct access to the abstracted underlay network.
Solution Approach 2:
The patent creates a copy of network traffic through synthetic probe flows that replicate actual application traffic patterns. These synthetic flows traverse the same SD-WAN paths and collect metrics at multiple hops, providing a mirrored view of network performance without interfering with real traffic. The copied metrics are then used to reconstruct end-to-end paths and performance characteristics.
2Measurement precision
If network monitoring applications collect data from multiple sources to provide comprehensive metrics, then measurement precision improves, but device complexity and difficulty of detecting and measuring increases
Solution Approach 1:
The patent segments the network monitoring function into distributed probes deployed at network edges and a centralized collection system. Each probe independently collects metrics from local traffic flows, reducing the complexity burden on any single device. The segmented architecture allows parallel data collection across multiple network nodes while maintaining centralized coordination for comprehensive metric aggregation.
Solution Approach 2:
The probes are designed to autonomously collect, process, and transmit network metrics without requiring complex configuration or manual intervention. The system self-manages probe deployment, data collection scheduling, and traffic correlation automatically, reducing operational complexity while maintaining comprehensive monitoring capabilities.
3Measurement precision
If synthetic traffic flows are used to traverse SD-WAN network for path visualization, then measurement precision of network paths improves, but loss of information about actual application traffic increases
Solution Approach 1:
The synthetic probe flows are designed to be universal in their functionality, capable of traversing various SD-WAN paths and collecting multiple types of metrics simultaneously. The same probe infrastructure serves both path visualization purposes and application traffic monitoring by analyzing synthetic flow characteristics that mirror actual traffic patterns. This multi-functionality allows the system to maintain information about both synthetic and actual traffic.
Data Source
AI summary
A device may query, by a path insight tool, an end-to-end network probe system to obtain test information about a traffic flow originating from an endpoint outside the SD-WAN by the end-to-end network probe system, the test information including one or more flow identifiers. A device may identify from the querying one or more SD-WAN overlay devices that carried the traffic flow having one or more flow identifiers, wherein one or more SD-WAN overlay devices are translated into an SD-WAN underlay path. A device may identify a flow identifier that identifies the traffic flow in the end-to-end network probe system, associating the test information about the traffic flow from the end-to-end network probe system with the statistics associated with the traffic flow obtained as the traffic flow traversed the SD-WAN underlay path. A device may generate an end-to-end visualization from the test information from the end-to-end network probe.