Abstract Topology for IP Network Path Calculation
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
In large and complex IP networks, the computational overhead for pathfinding algorithms increases due to the need to consider detailed topological information across multiple IGP areas and domains, which limits the ability to identify optimal TE paths that satisfy Service Level Agreement (SLA) constraints.
Innovation Solution
The method involves generating an abstract topology that represents domains and border routers in the actual IP network topology, determining measures of reachability and betweenness centralities for nodes, and modifying the graph by inserting temporary links to enhance reachability calculations. This abstract topology is then used to prune domains and adjust edge weights in the actual topology, optimizing pathfinding processes.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If detailed topological information across multiple IGP areas and domains is considered for pathfinding, then the accuracy of TE path identification is improved, but the computational overhead increases
Solution Approach 1:
The network topology is segmented into multiple IGP areas and domains, allowing pathfinding to be performed in a hierarchical manner. Routers within each area maintain detailed topological information locally, while inter-area pathfinding uses summarized route information. This segmentation enables accurate TE path identification within areas without requiring every router to process complete network-wide topological data, thus reducing computational overhead while maintaining pathfinding accuracy.
Solution Approach 2:
Area border routers and autonomous system border routers act as intermediaries that summarize and exchange routing information between different IGP areas and domains. These intermediary routers perform the computationally intensive pathfinding operations using detailed topological information, while interior routers use simplified routing tables. This intermediary approach allows accurate TE path calculation without requiring all routers to maintain and process complete detailed topological information, thereby reducing overall computational overhead.
2Reliability
If the full network topology is processed for path calculations, then optimal TE paths satisfying SLA constraints are identified, but the processing time increases
Solution Approach 1:
The system performs preliminary path calculations and constraint validation at area border routers before packets need to traverse multiple areas. By pre-computing inter-area paths and validating SLA constraints in advance, the system avoids time-consuming real-time calculations when packets are actually forwarded. This preliminary action ensures SLA compliance is maintained while reducing the time required for operational path calculations.
Solution Approach 2:
Different parts of the network perform different types of path calculations based on their local requirements. Interior routers perform simple intra-area pathfinding using detailed local topological information, while area border routers perform more complex inter-area path calculations. This local quality approach ensures that SLA constraints are properly validated at appropriate levels without requiring every router to perform time-consuming full-network path calculations, thus maintaining reliability while reducing processing time.
Data Source
Figure 1
Figure 2
Figure 3
AI summary
A processor generates an abstract topology that represents domains in an actual topology of a network and border routers that interconnect the domains in the actual topology. The processor prunes, based on the abstract topology, one or more of the domains from the actual topology during path calculations on the actual topology. In some cases, the processor determines measures of reachability for the nodes to indicate whether the domains represented by the nodes represent are on a path from a source domain to a destination domain. The processor can then selectively prune a relationship to at least one edge in the actual topology or modify weights of edges between the nodes in the actual topology based on the ranks of the nodes in the abstract topology.