Segment Routing Path Selection with Bandwidth Constraints
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing segment routing methods fail to efficiently find a minimum hop path in the segment graph that traverses the smallest number of links in the physical topology, leading to potential link overloading and inefficient bandwidth utilization.
Innovation Solution
A method is introduced to compute the segment graph, calculate bandwidth and number of links for each link, eliminate links with insufficient bandwidth, and select the shortest path using a subset of links to ensure a non-overlapping path with guaranteed bandwidth, utilizing Dijkstra's algorithm and segment routing principles.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Productivity
If conventional routing protocols like OSPF are used to determine paths, then packets are routed along shortest paths, but this leads to link overloading and hot-spots in the network
Solution Approach 1:
The patent applies segmentation by dividing the end-to-end path into multiple segments, where each segment is defined by a sequence of segment labels in the packet header. This allows the path to be composed of multiple shortest path fragments between intermediate nodes, enabling flexible traffic distribution while maintaining shortest path routing within each segment. The ingress node pushes a stack of segment labels to steer packets through specific intermediate nodes, thereby distributing traffic and avoiding link overloading.
2Productivity
If segment routing is used to distribute traffic and avoid hot-spots, then network utilization is improved, but finding the minimum hop path that traverses the smallest number of links becomes computationally complex
Solution Approach 1:
The patent introduces a new dimension to the path computation problem by considering not only the number of hops but also the number of links traversed in the physical topology. It defines a cost metric that combines both dimensions: cost = number of hops + penalty × number of links. This transformation converts the complex multi-constraint optimization problem into a single-objective shortest path problem that can be solved using modified Dijkstra's algorithm, thereby reducing computational complexity while still achieving traffic distribution goals.
3Adaptability or versatility
If explicit path routing with each hop specified in the packet header is used, then routing path control is maximized for service chaining, but per-flow state must be maintained at every intermediate node
Solution Approach 1:
The patent extracts the path control functionality from intermediate nodes and concentrates it at the ingress node. Instead of requiring per-flow state at every intermediate node, the ingress node computes the entire segment-routed path and encodes it as a stack of segment labels in the packet header. Intermediate nodes only need to perform simple label matching and popping operations without maintaining any per-flow state, thereby achieving explicit path control with minimal device complexity.
4Reliability
If a path is selected to guarantee minimum bandwidth, then bandwidth guarantee is achieved, but the path may not be the shortest path in terms of number of hops
Solution Approach 1:
The patent changes the parameter used for path selection from单纯的 number of hops to a composite cost metric that incorporates both number of hops and bandwidth availability. It assigns a cost to each link based on its residual bandwidth and uses this cost metric in the shortest path computation. This parameter transformation allows the algorithm to find paths that optimize the trade-off between path length and bandwidth guarantee, rather than simply selecting the shortest path or the highest bandwidth path in isolation.
Data Source
AI summary
Various embodiments relate to a non-transitory computer readable medium and method thereof for finding a minimum hop path in a segment graph traversing the least number of links in a physical topology, the method including receiving a connection request for a connection between a source node and a destination node, computing the segment graph, the segment graph having a plurality of links, computing a bandwidth for each of the plurality of links in the segment graph, computing the number of links for a shortest path (“N(q)”) for each of the plurality of links, eliminating each of the plurality of link with a bandwidth less than the minimum bandwidth and selecting the shortest path in the physical topology between the plurality of links.