A method and network node for realizing fast rerouting

Abstract

The invention discloses a method for realizing fast re-routing (FRR), which includes: building an FRR group according to a backup label switching path (LSP), setting marks of the FRR group to point to a next operation hop in the FRR group; modifying port information of the next hop in an original LSP into the marks of the FRR group; and setting the marks of the FRR group to point to the next protection hop in the FRR group when FRR switching is determined to be carried out. The invention further discloses a network node, by the aid of the method for implementing fast re-routing, many repeated operations during switching can be reduced, FRR switching speed is increased and self-routing management efficiency of the network node is improved.

Description

technical field [0001] The present invention relates to the technical field of Multi-Protocol Label Switching Traffic Engineering (MPLS TE, Multi-Protocol Label Switching Traffic Engineering), in particular to a method and a network node for realizing fast rerouting (FRR, Fast ReRoute). Background technique [0002] FRR is the technology for protecting links and network nodes in MPLS TE. Its basic principle is to establish a backup LSP for one or more Label Switching Paths (LSPs) to be protected in advance, and one or more of them to be protected The LSP can be called the active LSP, and the network node that manages the head node of the backup LSP is used as the repair node; when the repair node detects that a link failure occurs on the active LSP managed by itself or any network node fails, the repair node will The data traffic on the active LSP is switched to the established backup LSP. Wherein, the method of switching the data flow on the active LSP to the backup LSP is...

AI Technical Summary

Problems solved by technology

[0003] However, using the above method, when the repair node switches the data traffic on multiple active LSPs to the backup LSP, it needs to modify the next-hop port information corresponding to the active LSP to be switched one by one to the next-hop port information corresponding to the backup LSP. The port information of the jump will generate a large number of repeated operations, which will affect the switching speed

[0004] It can be seen that the FRR method in the prior art cannot reduce the number of repeated operations for modifying the next-hop port information of the primary LSP, which affects the speed of FRR switching, thereby affecting the efficiency of network nodes in managing their own routes

Images