Centralized LSP Selection Arbitrating QoS and Load
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Current methods for selecting a label switched path (LSP) in MPLS systems face challenges in managing conflicting constraints, particularly when network load increases due to information flooding, leading to inefficiencies in path calculation and inability to suitably control constraint conflicts.
Innovation Solution
An apparatus and method that analyze and digitize link performance information and fault history to determine standardized values, using a link weight managing unit to prioritize constraints based on administrative network policies, and an LSP selecting unit to choose paths that satisfy objective constraints and minimize expense using dynamic link weight values.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If Dijkstra's algorithm with QoS constraints is used for path selection, then path selection accuracy is improved, but network load increases due to information flooding
Solution Approach 1:
The patent extracts the path calculation function from distributed nodes and centralizes it in a network management server. This removes the information flooding burden from individual nodes while maintaining accurate path selection capability through centralized control that aggregates network status information efficiently.
Solution Approach 2:
The network management server acts as an intermediary between network nodes and path selection requirements. It collects network status information from nodes, processes constraint conflicts, and determines optimal paths, thereby mediating between distributed network operations and centralized path optimization without requiring nodes to exchange extensive information directly.
2Productivity
If centralized network management system is used to reduce information exchange load, then network efficiency is improved, but constraint conflict control capability deteriorates
Solution Approach 1:
The patent segments constraint handling into hierarchical levels: hard constraints (bandwidth, delay) are enforced strictly, while soft constraints (load balancing, stability) are used for optimization. This segmentation allows the centralized system to efficiently handle different constraint types with appropriate algorithms, reducing overall complexity.
Solution Approach 2:
The system dynamically adjusts link weight parameters based on network conditions and constraint priorities. By changing parameters like link weights and cost metrics according to real-time network status, the centralized management system can adapt to different scenarios without requiring complex decision logic at each node.
3Reliability
If multiple constraints are enforced simultaneously in path selection, then service quality is improved, but path selection complexity increases
Solution Approach 1:
The patent applies different constraint enforcement strategies to different parts of the path selection process. Hard constraints are applied locally to individual links during path evaluation, while soft constraints are applied globally to the overall path selection. This local quality approach allows multiple constraints to be enforced without uniformly increasing complexity throughout the entire system.
Data Source
AI summary
In an apparatus and method of selecting a label switched path (LSP) in a network management system, a status analyzing unit analyzes and digitizes performance information, fault occurrence, and fault history of a link. A link weight managing unit which determines link weight values using a predetermined weight calculation algorithm on second constraints that corresponds to qualitative information related to performance analysis of a path. An LSP selecting unit selects a path, from among paths on a path list, that satisfies first constraints which corresponds to objective information related to a path selection required by the administrative network policy, and, when there are several selected paths, selects a path that minimizes expense using the link weight values.