Cross-Domain Path Computation Using Interactive PCE Segmentation
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
In large-scale networks, existing methods for computing cross-domain separated paths often fail to obtain two separated paths due to the exclusion of paths that the first path passes through, leading to a low computation success rate and reliability issues.
Innovation Solution
The method involves interactive computation between path computation domains, where one domain adjusts its traffic engineering database based on path information and separated relationship information to compute a second path, including path information about boundary nodes that the first path passes through, thereby avoiding the exclusion of existing separated paths.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Reliability
If the first path is computed and its resources are excluded from the TED to compute the second path, then the separation constraint is satisfied, but the computation may fail even when separated paths exist
Solution Approach 1:
The patent segments the path computation process into two independent phases: first computing the shortest path without exclusion constraints, then computing the separated path by excluding only the first path's resources. This segmentation allows the system to explore multiple path options without being prematurely constrained, thereby increasing computation success rate while maintaining manageable complexity through structured computation steps.
Solution Approach 2:
The patent applies preliminary action by first computing the shortest path (first path) and storing its resource information before computing the second separated path. This preliminary computation provides the necessary path information for subsequent separation constraints without preventing the discovery of valid separated paths. The first path computation serves as a preparatory step that enables the second path computation to satisfy separation constraints while avoiding false failures.
2Ease of operation
If each node collects overall network topology through OSPF-TE equipment, then path computation can be performed, but control bandwidth resources are consumed and large memory capacity is required
Solution Approach 1:
The patent extracts the path computation function from individual nodes and consolidates it into a dedicated PCE (Path Computation Element). This extraction allows nodes to maintain only local topology information rather than collecting entire network topology, significantly reducing the memory capacity and control bandwidth requirements. The PCE performs complex path computation using extracted topology data from multiple nodes, achieving path computation capability without requiring each node to store and process the complete network topology.
3Quantity of substance
If a PCE is configured for each domain to perform path computation, then resource consumption is reduced, but interactive computation between multiple PCEs is required for cross-domain paths
Solution Approach 1:
The patent merges the path computation functions of multiple domain PCEs into a coordinated interactive computation process. When computing cross-domain paths, the source node's PCE and destination node's PCE exchange topology information and compute results collaboratively. This merging approach allows each PCE to maintain localized topology data (reducing individual resource requirements) while achieving comprehensive cross-domain path computation through coordinated interaction, effectively balancing resource efficiency with computation capability.
Data Source
Figure 1
Figure 2A
Figure 2B
AI summary
A method for obtaining a path, a path computation element (PCE), and a path computation system are provided. The method includes the following steps. At least two path computation domains between a path computation initial domain and a path computation termination domain perform interactive computation to obtain a path between two end nodes. At least one of the path computation domains obtains path information about a first path, computes and generates path information for computing a second path according to the obtained path information about the first path, and sends the generated path information to a next path computation domain. The path information for computing the second path includes path information about a path between two boundary nodes. One of the boundary nodes is a node that the first path passes through.