Routing Cost Normalization for Latency and Preference Trade-offs
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
In Interior Gateway Protocol (IGP) networks, changes to link costs or metrics intended to improve latency can adversely affect customer route preferences by altering the priority of routing pairs, potentially disrupting the network's ability to maintain traffic within specific geographic regions.
Innovation Solution
A computing device receives original and proposed routing information files, ranks routing pairs based on specified criteria, identifies changes in rank orders, and adjusts proposed costs to maintain the original rank order while confirming latency improvements, using a management terminal with components like a ranking module, latency monitor, and priority classifier to generate recommendations for network engineers.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Loss of time
If link costs are changed to improve latency, then network performance is improved, but customer route preferences are adversely affected
Solution Approach 1:
The system implements a feedback mechanism by comparing the rank order of routing pairs before and after cost changes. The routing pair rank comparator detects changes in routing priorities and feeds this information back to the cost normalization process, enabling automatic adjustment of costs to maintain customer route preferences while still achieving latency improvements.
Solution Approach 2:
The system dynamically adjusts link cost parameters based on detected changes in routing pair priorities. By modifying cost parameters selectively and iteratively, the system maintains the relative rank order of routing pairs (preserving customer preferences) while optimizing overall network latency performance.
2Loss of time
If link costs are adjusted to optimize network performance, then latency is reduced, but the priority of routing pairs changes
Solution Approach 1:
The system performs preliminary analysis by calculating and storing the original rank order of routing pairs before applying cost changes. This preliminary information is used to guide the cost adjustment process, ensuring that the final cost configuration maintains the desired routing pair priorities while achieving latency optimization.
Solution Approach 2:
The routing pair rank comparator continuously monitors changes in routing pair priorities resulting from cost adjustments. This feedback loop enables the system to detect and correct priority changes, maintaining stability in the routing pair composition while achieving performance improvements.
3Loss of time
If routing costs are modified to improve network performance, then latency improves, but geographic routing constraints are violated
Solution Approach 1:
The system converts the potential harm of geographic routing violations into a benefit by using the detected rank order changes as feedback signals. These signals guide selective cost adjustments that maintain geographic routing constraints while achieving latency improvements, turning what could be harmful violations into opportunities for optimized cost configuration.
Solution Approach 2:
The system selectively modifies link cost parameters based on their impact on geographic routing constraints. By analyzing which cost changes cause geographic violations and adjusting accordingly, the system maintains geographic routing integrity while achieving overall network performance optimization.
Data Source
AI summary
A computing device ranks multiple routing pairs based on a current routing configuration for weighted links in a network, and ranks the multiple routing pairs based on a proposed routing configuration, different than the current routing configuration, for the weighted links in the network. The computing device compares a current rank order of a first routing pair and a second routing pair in the multiple routing pairs with a proposed rank order of the first routing pair and the second routing pair. The computing device identifies a change between the current rank order and the proposed rank order and generates a data structure that indicates the identified change between the current rank order and the proposed rank order for the first routing pair and the second routing pair.