A data center network redundancy control device using a multi-chassis cluster system

Abstract

The invention relates to a data center network redundancy control device by adopting a multi-chassis cluster system, which is connected with the internet and comprises a multi-chassis cluster system, an access switch and an IDC entrust server, wherein the multi-chassis cluster system comprises at least two router clusters, the router clusters are mutually connected with the internet respectively according to a BGP protocol, the router clusters are mutually connected according to an OSPF protocol, and the access switch is respectively connected with the router clusters, the IDC entrust server and the internet. Compared with the prior art, the data center network redundancy control device by adopting the multi-chassis cluster system has the advantages that the topological structure is simple, shortcomings of the traditional STP (spanning tree protocol) can be overcome, communication reliability is improved and the like.

Description

technical field [0001] The invention relates to a network redundancy control device, in particular to a data center network redundancy control device using a multi-cassette cluster system. Background technique [0002] With the rapid development of Internet technology and the wider range of network applications, people have put forward higher requirements for information services, and these requirements have also put forward higher requirements for data center network platforms, and proper redundancy is urgently needed The network architecture can not only ensure the reliability of the network, but also meet the requirements of cost and energy consumption. [0003] At present, in the design of redundant network architecture in the IDC data center field, it is necessary to consider that if a node or a link in the network fails, it may lead to the isolation of users from core equipment or network services. Therefore, through redundant network architecture design Redundant bac...

AI Technical Summary

Problems solved by technology

[0004] 1) The spanning tree protocol achieves network redundancy through interception, learning, negotiation, blocking and other means of the switching network. The main link and the auxiliary link are a backup relationship. Only one of the two can be online at the same time. A loop will be generated when it is used, which will have a fatal impact on the network and make the network investment output low.

[0005] 2) After the main link is interrupted, an STP negotiation process is required to start the backup link, which takes a long time and is not conducive to the high reliability and high availability of the data center network;

[0006] The conventional redundant network architecture is that IDC managed servers, access switches, aggregation layer switches, routers, and the Internet are connected sequentially, which has the disadvantages of many devices, complex network topology, and complex maintenance and management.

[0007] Chinese patent 200610111302.1 discloses a method and device for preventing loops when RRPP and local STP networking fails to recover, and proposes a configuration message and a HELLO report for setting the faulty port of the transmission node on the faulty link and the temporary blocking time greater than STP The sending interval of the text, and the method of not blocking the STP configuration message when the faulty port is in the temporary blocking state, has the disadvantages of complex configuration, RRPP and STP protocol messages alternately manage the loop port on and off, etc.

Images