SRv6 Packet Path Tracing with Nested Header Groups
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Current hardware limitations restrict the number of midpoints that can be recorded along a packet path in path tracing implementations, preventing effective tracing in large diameter networks.
Innovation Solution
A method for generating a second header group that encapsulates the first header group when the first hop-by-hop header reaches maximum capacity, allowing for extended path tracing by appending a second hop-by-hop header and segment routing header, and updating the packet with new hop-by-hop information at each midpoint node.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Quantity of substance
If current path tracing implementations are used, then path tracing can be performed, but the number of midpoints that can be recorded is limited by hardware edit-depth capabilities
Solution Approach 1:
The patent divides the path tracing data structure into multiple stacks (first stack, second stack, third stack) organized in header groups. When the first stack reaches maximum capacity, a second header group with a second stack is generated to continue recording midpoints. This segmentation allows the system to record an extended number of midpoints beyond the hardware edit-depth limitation of a single stack.
Solution Approach 2:
The patent implements nesting by encapsulating the first header group (containing the first stack) within a second header group (containing the second stack). The second header group acts as a container that holds the first header group, allowing hierarchical organization of path tracing data across multiple stacks while maintaining a unified data structure.
2Duration of action of moving object
If the first header group reaches maximum capacity, then path tracing must stop, but extending path tracing requires additional header groups
Solution Approach 1:
The patent establishes predetermined maximum capacities for each stack (first maximum capacity, second maximum capacity) before path tracing begins. When a stack reaches its predetermined capacity, the system automatically triggers the generation of the next header group with its own stack. This preliminary capacity planning enables seamless extension of path tracing without requiring complex runtime decisions about when to expand.
3Loss of information
If more midpoints are recorded, then comprehensive path information is achieved, but hardware edit-depth limitations are exceeded
Solution Approach 1:
The patent transitions from a single-dimensional path recording approach (one stack) to a multi-dimensional approach by organizing multiple stacks (first, second, third stacks) across multiple header groups. This dimensional expansion allows the system to record comprehensive packet path information including all midpoints, source, and destination nodes while working within the constraints of hardware edit-depth capabilities for individual stacks.
Data Source
AI summary
The present disclosure provides systems and methods for enabling packet path tracing (PT) in large diameter networks that would otherwise exceed a hop limit imposed by hardware edit-depth limitations. Upon determining that a Midpoint Compressed Data (MCD) stack of a packet header is full, a PT Encapsulation and Copy (PEC) enabled midpoint node can apply an encapsulation protocol that generates a new header grouping that encapsulates a previous header grouping therein, copies information from the previous header grouping into the new header grouping, and adds a new MCD stack to the packet header for further population of path tracing data and to ensure that the packet remains along the intended path. The encapsulation protocol ensures that the new header grouping remains within the edit-depth of each midpoint node encountered by the packet. The path tracing information can be combined at a controller resulting in full path information for the packet.