Data Plane Backup Path Validation in Segment Routing
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Current network technologies face challenges in validating backup paths in Segment Routing networks without involving the control plane, which is essential for ensuring network uptime and service level agreements.
Innovation Solution
The implementation of data plane-based backup path validation techniques using a domain-wide unique label (Backup Context Segment Identifier) allows for the creation of a Backup Context Table and the use of probe packets with data plane instruction labels to validate backup paths without control plane intervention, leveraging Segment Routing and OAM server functionality.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Reliability
If control plane intervention is used for backup path validation, then validation can be performed, but network complexity and validation overhead increase
Solution Approach 1:
The patent extracts the backup path validation function from the control plane and implements it purely in the data plane. Probe packets carry segment routing headers that encode backup path instructions, allowing validation to occur through data plane packet forwarding without control plane message exchanges or state maintenance at transit nodes.
Solution Approach 2:
The validation system is self-service in that probe packets automatically traverse the backup path and return to the source node without requiring control plane intervention at intermediate nodes. The segment routing headers in the probe packets contain all necessary instructions for nodes to forward packets along the backup path based on their forwarding tables.
2Productivity
If data plane-based validation is implemented, then validation overhead is reduced, but mechanisms for backup path identification and traversal must be established
Solution Approach 1:
The patent changes the parameter representation by encoding backup path identification information directly in the segment routing header parameters. The segment list in the probe packet contains segment identifiers that correspond to nodes on the backup path, and the segment routing header format is modified to include context identifiers that reference pre-computed backup path information stored in forwarding tables.
3Speed
If backup path information is stored at each node, then fast backup path switching is enabled, but memory requirements and information storage overhead increase
Solution Approach 1:
The forwarding tables at each node are designed to be multi-functional, serving both primary path forwarding and backup path validation purposes. The same forwarding table that routes normal traffic also contains the backup path information needed for validation, eliminating the need for separate storage structures and reducing overall memory requirements.
Data Source
AI summary
In one embodiment, a server has connectivity to a network that includes a plurality of nodes. The server stores a context identifier that is associated with backup path information stored at a particular node of the plurality of nodes, the backup path information indicating a backup path in the network. The server sends to the particular node a probe packet having a data plane instruction label that includes an identifier of the particular node, the context identifier and server identifier information. The context identifier in the probe packet causes the particular node to retrieve the backup path information and send the probe packet to a next hop node in the backup path based on the backup path information. The server determines whether the probe packet is received at the server after traversing the backup path, and validates the backup path for the particular node accordingly.