A cloud data center disaster recovery system and method based on data distribution backup

Abstract

The invention discloses a cloud data center disaster recovery system and method based on data distribution backup, and relates to the field of cloud data backup and recovery. The method comprises a main data center and a standby data center, the main data center comprises a first host device and a first storage device, the first host device is used for storing management data, and the first storage device is used for storing business data. The standby data center comprises a second host device and a second storage device, wherein the first host device and the second host device are connected through an LAN management network, and the first storage device and the second storage device are connected through an optical fiber. The management data is the virtual machine configuration information and comprises the volume information, the network information and the virtual machine information; the service data is the virtual machine hard disk data, and the virtual machine hard disk data is LUN data corresponding to a virtual machine volume. According to the invention, the data backup synchronization and data recovery time can be reduced.

Description

technical field [0001] The invention relates to the field of data backup and recovery of cloud data, in particular to a cloud data center disaster recovery system and method based on data split backup. Background technique [0002] When building a cloud data center, it is often necessary to perform disaster recovery backup on the data center to avoid immeasurable losses caused by data loss and service interruption in the data center due to disasters. Existing cloud data centers generally use backup software and agents installed on hosts to manage the network based on LAN (Local Area Network) and copy data to remote data centers for backup. For recovery, specifically, make a snapshot of the data in the primary data center, package and compress it, and remotely copy it to the storage in the standby data center through the LAN management network. When a disaster occurs, restore the data in the standby data center by decompressing and merging the snapshots . Because it passes ...

AI Technical Summary

Problems solved by technology

Because it passes through the LAN management network, the bandwidth of data transmission is limited, and it will occupy the normal management network bandwidth. At the same time, the data link is long. When data is backed up, the data needs to be transferred from the storage device of the main center to the main center through the storage network of the main center. The host, then reaches the host in the backup center through the LAN management network of the master and backup centers, and then reaches the storage device in the backup center through the storage network of the backup center, so the copy speed is slow and the delay is large, and the RPO (Recovery Point Objective) of disaster recovery and RTO (Recovery Time Objective, recovery time objective) will be large, and it is difficult to meet the disaster recovery scenarios with small delay requirements

Images