Routing Area Abstraction with Representation Nodes to Reduce Storage
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Inclusion of entire network topology information in link state databases increases storage burden and can lead to sub-optimal convergence times in dense network topologies due to flooding overhead.
Innovation Solution
Abstract a network area by using an area representation node, where network devices in a second area advertise adjacency to this node instead of specific devices in the first area, reducing the need to store the entire topology in their link state databases.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If entire network topology information is included in link state databases, then routing accuracy is improved, but storage burden increases
Solution Approach 1:
The network topology is segmented into hierarchical levels (Level 1 areas and Level 2 areas). Level 1 areas maintain complete topology information locally, while Level 2 areas use simplified representations (summary LSA) of Level 1 areas. This segmentation allows routing accuracy to be maintained where needed while reducing storage burden in other areas.
Solution Approach 2:
Complete topology information is extracted from Level 1 areas and selectively distributed to Level 2 areas through summary LSA. Level 2 areas take out only the essential routing information needed for inter-area routing, rather than storing entire Level 1 topology details, thus reducing storage requirements while maintaining routing functionality.
2Reliability
If flooding is used to distribute topology information, then routing information is kept current, but convergence time increases in dense topologies
Solution Approach 1:
The flooding scope is segmented and limited to Level 1 areas only. Summary LSA are generated at Level 1 area boundaries and distributed to Level 2 areas without triggering full flooding. This segmentation reduces the flooding overhead in dense topologies while maintaining routing information currency through targeted updates at hierarchical boundaries.
3Measurement precision
If detailed topology information is stored, then routing decisions are more accurate, but device performance decreases
Solution Approach 1:
Different areas have different qualities of topology information stored locally. Level 1 area devices store complete detailed topology information for accurate intra-area routing decisions. Level 2 area devices store simplified summary representations for efficient inter-area routing. This local quality differentiation maintains routing accuracy where detailed information is available while improving device performance where simplified information suffices.
Data Source
Figure 1
Figure 2
Figure 3
AI summary
Embodiments of the invention may relate to a method for routing protocol area abstraction. The method may include electing an area leader from among network devices; generating, by the area leader, an area representation node identifier associated with the first area; distributing, by the area leader, the area representation node identifier to area edge devices; receiving, from the area edge devices, second area link state packets (LSPs); generating, by the area leader and using the second area LSPs, an area representation node LSP that includes the area representation node identifier and area neighbor adjacencies; and distributing, by the area leader, the area representation node LSP to a plurality of network devices in a second area of the network. In response to receiving a copy of the area representation node LSP, each of the network devices in the second area may advertise an adjacency to an area representation node.