Switchpath Calculation in Label Switched Networks
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
In label switched networks, the existing methods for determining data stream switchpaths often lead to congestion due to reliance on single criteria like path length, which neglects other pertinent factors such as bandwidth, and require extensive calculation time, limiting reliability and resource suitability.
Innovation Solution
A device that calculates switchpaths by considering multiple criteria and constraints, using a table of correspondences for service data and network topology, assigning interest values to paths based on performance, and selecting paths that optimize these criteria while ensuring connectivity and resource reservation.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Device complexity
If a single selection criterion (such as path length) is used to calculate switched paths, then the calculation process is simple, but traffic congestion occurs because most traffic is directed along the same path while other suitable paths remain underused
Solution Approach 1:
The patent segments the path selection process by dividing criteria into multiple independent dimensions (length, bandwidth, number of hops, etc.) and evaluates paths based on segmented criteria rather than a single unified metric. This allows traffic to be distributed across multiple paths based on different criteria segments, preventing congestion while maintaining manageable calculation complexity through modular evaluation.
Solution Approach 2:
The patent transitions from one-dimensional path selection (single criterion) to multi-dimensional path selection by introducing additional evaluation dimensions such as bandwidth availability, number of hops, and administrative weight. This dimensional expansion enables the system to identify and utilize multiple suitable paths that would be invisible under single-criterion evaluation, thereby distributing traffic and preventing congestion.
2Reliability
If multiple criteria are considered for path calculation, then path selection optimality improves, but calculation time increases significantly
Solution Approach 1:
The patent performs preliminary actions by pre-calculating and storing path metrics for each criterion (length, bandwidth, hops, etc.) in advance. When path selection is needed, the system retrieves these pre-computed values and combines them according to weighted criteria, rather than performing complete multi-criterion calculations from scratch. This preliminary preparation significantly reduces real-time calculation time while maintaining optimal path selection.
Solution Approach 2:
The patent changes parameters by introducing weighted coefficients for each criterion that can be adjusted to balance optimality and calculation speed. By modifying the weight parameters assigned to different criteria, the system can optimize the trade-off between path selection quality and computation time, allowing flexible adaptation to different network conditions and requirements.
3Reliability
If constraints are applied to ensure QoS and ToS requirements, then service quality improves, but the choice of available paths is limited and congestion may be aggravated
Solution Approach 1:
The patent applies local quality by evaluating different path segments against specific QoS and ToS constraints locally rather than applying uniform filtering across all paths. Each path is assessed against relevant constraints at critical points (such as bandwidth requirements at bottleneck links, or delay constraints at specific hops), allowing paths to satisfy local quality requirements while maintaining overall flexibility in route selection.
Solution Approach 2:
The patent introduces dynamics by making constraint application adaptive rather than static. The system dynamically adjusts which constraints are applied and with what strictness based on current network conditions, traffic patterns, and service requirements. This dynamic approach allows the path selection to adaptively balance between meeting QoS/ToS requirements and maintaining path selection flexibility, preventing unnecessary congestion from overly rigid constraint enforcement.
Data Source
AI summary
A device (D) dedicated to determining data stream switchpath(s) in a label switched network comprises a table (Tn1) of correspondences between sets of service data and information data representative of at least two criteria, a descriptive structure (Tn2) containing information data representative of the state of utilization and of the topology of the network, and a processing module (Pn) adapted i) to receive a path set-up request containing a set of service data associated with a stream to be switched, to determine in said table (Tn1) the criteria associated with said set of service data associated with the stream, ii) to ensure the connectivity of the network nodes on the basis of information data stored in said descriptive structure (Tn2), iii) to calculate from among the nodes of the network possible paths between a departure node (LER1) and a destination node (LER2) taking account of the criteria, and then to deduce an ideal solution from performances of the possible paths on the criteria, iv) to assign each possible path an interest value taking account of the ideal solution and then to classify said possible paths taking account of their respective interest values, and v) to select one of the classified possible paths and then to associate with the stream to be switched a label representative of the selected path so that said labeled stream is switched to the destination node.