Method for realizing port fast reroute and device

Abstract

The invention discloses a method for realizing port fast reroute; a route start port switching marked position is set in a port list of a fast reroute entry; the method comprises the following steps: when a primary path is detected to lose efficacy, the value of the start port switching marked position in the port list corresponding to the start port of the primary path is set into the value that the primary path loses efficacy. The invention further discloses a device for realizing port fast reroute; by adopting the invention, the switching speed of the primary path is improved.

Description

technical field [0001] The present invention relates to fast rerouting (FRR, Fast ReRoute) technology, in particular to a method and device for realizing port fast rerouting. Background technique [0002] The traditional Internet Protocol (IP) network service model is a "best-effort" service model. With the further development of network services, the reliability of the IP network as a unified multi-service bearer must reach the level of traditional telecommunication networks , such as the speed of protection switching is less than 50ms, in order to meet the needs of the carrier-class business. [0003] The FRR technology is a technology used to quickly protect a local network, and it is generally deployed in a network that requires high reliability. When a link or node fails in the network, the FRR can quickly switch to the backup path, so that the data service is less affected. [0004] Currently, IP FRR, Label Distribution Protocol (LDP, Label Distribution Protocol) FRR...

AI Technical Summary

Problems solved by technology

[0022] It can be seen from the above process that the existing FRR switching method needs to rely on the FRR group information issued by the support about the next hop of the failed active path before performing the active / standby path switchover, and the switching needs to be performed for each FRR Group switching, the process is too cumbersome, and the switching time takes up too much

Especially when there are many FRR groups, the overhead is greater, which will inevitably affect the time to complete the handover. For example, if 100 FRR groups share one active link or a link in the active path, the existing FRR handover method Each FRR group needs to be switched, that is, 100 operations are required to complete all the switches

[0023] In addition, the invention patent application with the application number 200410056187.3 discloses a method for implementing rerouting in an IP network. This method implements switching by modifying the port table, but the method for detecting whether the port status is valid in this application is limited by the scanning of the hardware itself. Detection period, generally, the scanning detection period of a physical port is greater than 100ms. If the period is too small, it will occupy the resources of the CPU. In this way, the time spent from the detection of the port failure to the switch is far beyond 50ms, so it cannot meet the requirements of telecommunications. level business needs

Images