Cross-Domain Path Identification Compression via Binding SIDs
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing methods for determining cross-domain TE paths result in excessively long identification information, leading to forwarding inefficiencies and MTU problems due to deep label stacks, and sub-PCEs may not provide necessary path segment information to parent PCEs.
Innovation Solution
A method that compresses identification information for cross-domain TE paths by instructing sub-PCEs to create tunnels with Binding SIDs, allowing for the combination of compressed information into a shorter SID list, which is then used for packet forwarding.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If the complete TE path is represented by an adjacency SID + Peer-Adj SID list, then the path can be accurately identified, but the SID list becomes excessively long, causing deep label stacks that exceed device label encapsulation capability and reduce forwarding efficiency
Solution Approach 1:
The patent divides the complete TE path into multiple path segments, each corresponding to a domain. Instead of representing the entire path with a single long SID list, each domain's sub-PCE computes and returns a separate path segment. The parent PCE then splices these segmented paths together, allowing each segment to have a manageable SID list length while maintaining complete path identification capability.
Solution Approach 2:
The patent introduces a hierarchical dimension to the path representation. Rather than a flat, single-level SID list, the path is represented hierarchically with domain-level segmentation. Each domain contributes its own SID list for its segment, and these are combined in a hierarchical manner, effectively adding a domain-level dimension to the path identification structure.
2Loss of information
If sub-PCEs return path segment information to the parent PCE, then complete path information can be obtained, but sub-PCEs may not necessarily provide the information according to local strategy, leading to incomplete path data
Solution Approach 1:
The patent implements a feedback mechanism where the parent PCE sends requests to sub-PCEs for path segment computation, and sub-PCEs return their computed segments. The parent PCE then splices these segments together and can request additional information or corrections if needed. This feedback loop ensures that complete path information is obtained while allowing sub-PCEs to apply their local strategies in computing segments.
Solution Approach 2:
The patent creates a universal interface between parent PCE and sub-PCEs through standardized path computation requests and responses. This universal mechanism allows sub-PCEs to maintain their local strategies for computing path segments while ensuring that all necessary information is eventually gathered by the parent PCE through the standardized interface.
3Extent of automation
If a hierarchical PCE architecture is used to compute cross-domain TE paths, then path computation can be distributed and scaled, but the coordination between parent PCE and sub-PCEs increases system complexity
Solution Approach 1:
The patent segments the path computation function across multiple PCEs in a hierarchical architecture. The parent PCE handles domain-level topology and coordinates between domains, while sub-PCEs handle intra-domain path computation. This segmentation distributes the computational load and allows scaling, while the clear division of responsibilities reduces coordination complexity compared to a fully centralized approach.
Data Source
Figure 1~2
Figure 3~5
Figure 6~7
AI summary
Provided are a method and apparatus for determining identification information about a cross-domain path, and a storage medium. The method comprises: receiving a path computation request from a first sub-path computation element (PCE), wherein the path computation request carries a compression identifier, for identifying the execution of compression on a path segment, and a path set-up type (PST); after identification information about a path from a source node to a destination node is acquired, instructing, according to the compression identifier and the PST, a sub-PCE corresponding to each path segment to compress identification information about the respective corresponding path segment; and receiving the compressed identification information about each path segment which is returned by the sub-PCE corresponding to the path segment, and according to the identification information about the path, combining the compressed identification information about the path segments into the compressed identification information about the path.