Constrained Shortest Path Determination in Networks
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
As computer networks grow, determining the shortest path between nodes becomes increasingly time-consuming and costly due to the need for constrained path computations, which are critical for routing data efficiently, especially in scenarios like real-time data flows that require bandwidth guarantees and low latency.
Innovation Solution
A method for determining multiple shortest paths in a network from a source node to destination nodes involves generating a shortest path tree and a candidate list, where each candidate includes a working node, minimum cost, maximum available bandwidth, and previous hop node, while adhering to path constraints such as bandwidth, ensuring efficient computation by updating available bandwidth and pruning paths that do not meet constraints.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Reliability
If constrained shortest path computations are performed in traditional networks, then routing accuracy and service quality are improved, but computational time and processing cost increase significantly
Solution Approach 1:
The patent pre-computes and stores shortest path information in routing tables before actual data transmission occurs. When a data packet needs to be routed, the router simply looks up the pre-computed path in the routing table rather than performing constrained shortest path computation in real-time. This preliminary action resolves the contradiction by shifting computational work from the critical path (increasing reliability) to an offline preprocessing stage (reducing time loss).
Solution Approach 2:
The patent creates simplified copies of the network topology and path information in routing tables that can be quickly queried. Instead of performing complex constrained shortest path computations on the actual network graph, routers use these pre-generated copies containing essential routing information. This copying approach maintains routing accuracy while dramatically reducing computational time during actual packet forwarding.
2Adaptability or versatility
If the number of nodes in the network increases, then network capacity and coverage are improved, but the number of path computations and processing complexity increase
Solution Approach 1:
The patent divides the network into autonomous systems or domains, each maintaining its own routing tables independently. Instead of performing path computations across the entire large-scale network, routers only need to compute and store paths within their local domain. This segmentation reduces processing complexity while maintaining the ability to handle large networks through hierarchical routing between domains.
Solution Approach 2:
The patent performs path computation and routing table generation in advance for each autonomous system, before actual data transmission occurs. This preliminary action allows complex constrained shortest path computations to be performed offline when computational resources are abundant, while online packet forwarding uses simple table lookups. This resolves the contradiction by enabling large network capacity through pre-computation while keeping real-time processing complexity low.
3Reliability
If constrained path computations are centralized in a single SDN controller, then path optimization and service quality are improved, but the controller's processing load and single point of failure risk increase
Solution Approach 1:
The patent distributes the centralized SDN controller functionality across multiple controller instances or divides the network into multiple autonomous systems, each with its own controller. Instead of one controller handling all path computations for the entire network, the processing load is segmented and distributed. This reduces the processing load on individual controllers while maintaining optimized path computation through localized control planes.
Solution Approach 2:
The SDN controllers perform constrained shortest path computation and generate routing tables in advance, before actual data traffic occurs. This preliminary action shifts the heavy computational workload from real-time packet processing to offline path computation. The controllers can perform complex computations when traffic load is lower, then use the pre-computed routing tables for fast online packet forwarding, reducing real-time processing load while maintaining service quality.
Data Source
AI summary
A method of determining a plurality of shortest paths in a network from a source node to respective destination nodes comprises: accessing, by one or more processors, for each of the plurality of shortest paths, a path constraint; generating, by the one or more processors, a shortest path tree (SPT) and a candidate list, wherein each candidate of the candidate list comprises a working node, a minimum cost for a path to the working node from the source node, a maximum available bandwidth for the path to the working node from the source node, and a previous hop node; and determining, by the one or more processors, the plurality of shortest paths in the network from the source node to the respective destination nodes based on the SPT, each determined shortest path meeting the path constraint.