In-situ Data Packet Detection for Network Path Telemetry
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Conventional OAM detection methods in networks fail to ensure consistent transmission paths for data flows and cannot perform per-packet detection, limiting the identification of data transmission paths and network performance measurement.
Innovation Solution
A data packet detection method that encapsulates detection information into data packets, allowing network devices to collect and forward information in-situ while maintaining packet forwarding performance, enabling in-situ flow information telemetry and transmission path detection.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Reliability
If conventional OAM detection packets are used, then transmission path detection is provided, but the transmission path of the OAM detection packet cannot be ensured to be consistent with the transmission path of the data flow
Solution Approach 1:
The patent merges the detection function with the data packet itself by encapsulating detection information within the data packet. This ensures that the detection packet travels along the exact same path as the data flow, eliminating path inconsistency issues. The detection information is embedded in the packet header or payload, making the detection packet indistinguishable from regular data packets in terms of routing.
Solution Approach 2:
The patent introduces detection information as an intermediary element that mediates between the data packet and the detection function. This detection information acts as a marker or tag that enables path tracking without altering the fundamental data packet structure or routing behavior, ensuring consistency while maintaining detection capability.
2Adaptability or versatility
If IP FPM technology is used to color data packets, then per-flow detection is implemented, but per-packet detection cannot be implemented and data transmission path identification is limited
Solution Approach 1:
The patent segments the detection information into discrete fields within the packet structure, including packet identifier, sequence number, detection domain identifier, and collection information. This segmentation enables per-packet detection by allowing each packet to carry its own unique detection metadata, while the structured format keeps processing complexity manageable through standardized field definitions.
Solution Approach 2:
The patent applies local quality by making detection information specific to each packet's context. Each packet carries its own identifier and sequence number, allowing individual packet tracking. The collection information is tailored to the specific detection task type, enabling flexible, packet-level detection without requiring complex system-wide changes.
3Productivity
If detection information is encapsulated into data packets for in-situ collection, then packet forwarding performance is maintained, but additional processing overhead is introduced
Solution Approach 1:
The patent performs preliminary action by pre-defining the detection information structure and encapsulation format. The packet identifier, sequence number, and collection information fields are established in advance according to the detection task type. This pre-configuration enables efficient processing during packet forwarding, as devices can directly extract and process the pre-structured detection information without complex real-time analysis.
Data Source
AI summary
A data packet detection method, a device, and a system are disclosed. The method includes: receiving first control information sent by a controller; receiving a first data packet sent by a previous-hop network device of a first network device, where the first data packet includes first detection information, and the first detection information includes a first detection node identifier, a first sequence number, and first collection information; determining, based on the first collection information, first collected data corresponding to the first collection information, and updating the first detection node identifier; and sending, to a next-hop network device of the first network device, the first data packet carrying the updated first detection information. This implements information telemetry on a data packet on a transmission path.