Pre-computing Diverse Paths for Label Switched Path Routing
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
In large computer networks, constraint-based path computation for multiple Label Switched Paths (LSPs) is complex, time-consuming, and resource-intensive, especially as available bandwidth is reduced with each LSP loaded, leading to network congestion and slow convergence in case of failures.
Innovation Solution
A path computation device computes a result set of multiple paths between nodes using recursive or heuristic algorithms to identify locally-optimal paths for a cost function, allowing for pre-computation of alternate paths to distribute traffic evenly and quickly reroute in case of failures, thereby reducing the need for real-time recalculation.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Reliability
If constraint-based path computation is performed for multiple LSPs in large networks, then routing decisions can be made with bandwidth and failure constraints, but the computation becomes complex, time-consuming, and resource-intensive
Solution Approach 1:
The patent pre-computes multiple candidate paths (N paths) between node pairs and stores them in advance. When a routing decision is needed, the system selects from pre-computed paths rather than computing new paths in real-time. This preliminary computation of multiple paths resolves the contradiction by making the complex computation happen once during path setup rather than repeatedly during routing decisions.
Solution Approach 2:
The patent segments the path computation process into two phases: (1) pre-computation phase where multiple candidate paths are calculated and stored, and (2) routing phase where paths are selected from the pre-computed set. This segmentation separates the complex computational work from the quick routing decisions, reducing real-time computation complexity while maintaining routing accuracy.
2Reliability
If multiple LSPs are loaded sequentially with constraint-based path computation, then bandwidth constraints are enforced, but available bandwidth is reduced with each LSP loaded leading to network congestion
Solution Approach 1:
The system pre-computes multiple candidate paths for each LSP request and stores them in advance. When LSPs are loaded sequentially, the pre-computed paths are selected from the candidate set rather than computing new paths each time. This allows the system to enforce bandwidth constraints while avoiding the cumulative computational overhead that would otherwise reduce network productivity.
3Adaptability or versatility
If real-time path computation is performed for each LSP routing decision, then current network conditions are considered, but convergence time after failures increases due to slow recalculation
Solution Approach 1:
Multiple candidate paths are pre-computed and stored before failures occur. When a failure happens, the system can quickly switch to an alternative pre-computed path without waiting for recalculation. This preliminary preparation of backup paths dramatically reduces convergence time while still adapting to failure conditions through rapid path selection from the pre-computed set.
Solution Approach 2:
The patent separates path computation into pre-computation (offline) and path selection (online) phases. This segmentation allows the system to respond quickly to failures by selecting from pre-computed paths rather than performing real-time recalculation, thus reducing convergence time while maintaining adaptability to network conditions.
4Device complexity
If a single path is used for each LSP, then path computation is simple, but network congestion occurs and traffic cannot be distributed evenly
Solution Approach 1:
The system pre-computes multiple candidate paths (N paths) for each node pair and stores them in advance. When routing decisions are made, traffic can be distributed across multiple pre-computed paths rather than using a single path. This preliminary computation of multiple paths enables even traffic distribution without significantly increasing real-time computation complexity.
Data Source
Figure 1
Figure 2
Figure 3
AI summary
In general, techniques are described for identifying a result set of multiple paths through a network for one or more label switched paths between a source node and a destination node. In some examples, the identified paths are computed to be broadly separate to avoid overlapping network elements. A device that routes the label switched paths to the network may select from the result set of multiple paths to route each of the label switched paths. In response to detecting a failure of a network element along the routed path for a label switched path, the device may select a new path from the result set that includes already-identified paths.