Topology-Based Link State Consolidation for Network Routing
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing network elements face challenges in efficiently estimating and storing network traffic state information, particularly in large-scale networks with many ports and nodes, due to high memory, processing power, and communication overhead requirements.
Innovation Solution
The method involves identifying groups of destination nodes reachable over common network links, estimating performance metrics for these groups, and storing them in a Forwarding Database, rather than per individual node, to reduce memory and computational needs, and selecting optimal network paths based on these metrics for packet forwarding.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If network elements estimate and store traffic state information per individual destination node, then routing precision is improved, but memory space and computational power requirements increase significantly
Solution Approach 1:
The patent merges traffic state information for multiple destination nodes into aggregate group-level metrics. Instead of maintaining separate estimates for each individual node, the system consolidates data by grouping nodes and estimating performance metrics (such as latency or load) for the entire group, thereby reducing memory requirements while preserving sufficient routing information.
Solution Approach 2:
The patent segments the set of destination nodes into distinct groups based on shared characteristics or network topology. By dividing the large set of individual nodes into smaller manageable groups, the system can estimate traffic state for each group separately, reducing the overall computational and memory burden compared to handling every node individually.
2Measurement precision
If network elements estimate and store traffic state information per individual destination node, then routing decision accuracy is improved, but processing power requirements increase
Solution Approach 1:
The patent combines routing decisions for multiple destination nodes into a single group-level decision. By estimating performance metrics for groups rather than individuals, the system reduces the number of separate calculations and comparisons needed, thereby lowering processing power consumption while maintaining adequate routing accuracy for all nodes within the group.
Solution Approach 2:
The patent segments routing decisions into group-level operations rather than node-level operations. This segmentation reduces the total number of decision-making instances from N (number of nodes) to G (number of groups, where G < N), significantly reducing the processing power required for routing table updates and path selection.
3Productivity
If network elements maintain detailed traffic state information for all destination nodes, then adaptive routing performance is improved, but communication overhead increases
Solution Approach 1:
The patent merges communication traffic related to state information exchange by aggregating data at the group level. Instead of each node reporting individual state information, nodes within a group can collectively contribute to group-level metrics, reducing the volume of communication traffic and overhead while still enabling effective adaptive routing based on the consolidated information.
4Measurement precision
If network elements use traditional per-node state estimation, then routing accuracy is maintained, but device complexity increases
Solution Approach 1:
The patent merges multiple individual node data structures into fewer group-level data structures. By consolidating the information representation from N individual node entries to G group entries (where G < N), the system simplifies the overall data structure complexity, reduces memory allocation overhead, and makes the routing table more manageable while preserving essential routing accuracy.
Data Source
AI summary
A method in a network element that forwards packets to destination nodes includes identifying groups of the destination nodes. Respective performance metrics of multiple different candidate network paths, over which the destination nodes in a given group are reachable from the network element, are estimated jointly for all the destination nodes in the given group. A network path is selected from among the candidate network paths based on the estimated performance metrics. The packets addressed to the destination nodes in the given group are forwarded over the selected network path.