MPLS TE Tunnel Switching for Network Resource Convergence
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
MPLS TE technology deployed end-to-end lacks bandwidth convergence ability, leading to an N-square problem of tunnel numbers as network nodes increase, causing a heavy load on network core nodes and inefficient resource utilization.
Innovation Solution
Implementing a method and node device that enables MPLS TE tunnel switching by configuring a Border Gateway Protocol-based route reflector to modify routing and forwarding information, allowing tunnel switching between different MPLS TE tunnels, thereby partitioning the network into layered areas with intra-area fully-meshed tunnels and inter-area tunnel switching.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Reliability
If MPLS TE tunnels are deployed end-to-end between any two nodes, then QOS and reliability guarantee is improved, but the tunnel number increases N-square causing heavy load on network core nodes
Solution Approach 1:
The patent segments the flat end-to-end tunnel architecture into a hierarchical structure with area head nodes and ordinary nodes. Tunnels are established within areas rather than between all node pairs, reducing the N-square tunnel proliferation. Area head nodes aggregate traffic and establish inter-area tunnels, creating a layered topology that maintains reliability while reducing overall tunnel count.
Solution Approach 2:
The patent introduces a hierarchical dimension to the network architecture, transforming the flat two-dimensional node-to-node tunnel structure into a three-dimensional hierarchical structure with multiple levels (area head nodes and ordinary nodes) and layers (intra-area and inter-area tunnels). This dimensional change enables bandwidth convergence at the area head node level while maintaining end-to-end QOS guarantees.
2Reliability
If MPLS TE tunnels with constant bandwidth are deployed, then bandwidth resource reservation is improved, but bandwidth convergence ability is lost leading to inefficient resource utilization
Solution Approach 1:
The patent segments bandwidth management into two levels: intra-area bandwidth reservation maintained by MPLS TE for reliability, and inter-area bandwidth aggregation at area head nodes enabling convergence. Ordinary nodes preserve fixed bandwidth guarantees within their area, while area head nodes perform bandwidth convergence for inter-area traffic, combining reservation reliability with adaptability.
Solution Approach 2:
Area head nodes serve as intermediary entities between ordinary nodes and the core network. They aggregate bandwidth requests from multiple ordinary nodes, perform bandwidth convergence, and establish inter-area tunnels. This intermediary role enables both bandwidth reservation (through MPLS TE at ordinary node level) and bandwidth convergence (at area head node level), resolving the contradiction between fixed bandwidth and adaptability.
3Productivity
If the network is transformed into a layered structure with tunnel switching, then bandwidth resource utilization is improved, but routing and forwarding information complexity increases
Solution Approach 1:
The patent segments routing and forwarding information management into two levels: intra-area routing maintained by IGP for simplicity, and inter-area routing managed through area head nodes. Ordinary nodes only need IGP routing within their area, while area head nodes handle inter-area routing and tunnel switching. This segmentation reduces routing complexity at ordinary nodes while enabling efficient bandwidth utilization through hierarchical tunnel switching.
Data Source
Figure 1~2
Figure 3
AI summary
The core idea of the method for transferring a data message on the TE tunnel and a node device is to determine the transit node needed to switch the MPLS TE tunnel, to set a routing forward information by which the different MPLS TE tunnels switch the tunnels and to switch and transmit the data message in the different MPLS TE tunnels by the transit node according to the routing forward information. The sum of the whole network edge band width resource is higher than the band width resource in the core of network, the network resource is converged layer by layer, the problem of N square amount of the tunnels is avoided, and the pressure on the network core node caused by a great deal of tunnels and the state renewing pressure of the RSVP-TE are alleviated; the method for setting up a route forwarding information for switching the MPLS TE tunnels is simple and easy to realize; thereby realizing the aim that the requirement for the network resource converged layer by layer is satisfied, the utilization of the band width resource and the extensibility of network are increased.