![<?echo $_SERVER['SERVER_NAME'];?>](/template/twentyseventeen/skin/images/header.jpg)
Disaster recovery backup is to establish and maintain a backup storage system in different places, and use geographic separation to ensure the ability of the system and data to resist catastrophic events.
According to the degree of resistance of the disaster recovery system to disasters, it can be divided into data disaster recovery and application disaster recovery. Data disaster tolerance refers to the establishment of a remote data system that replicates key application data in real time in the local system. When a disaster occurs, the remote system can quickly replace the local system to ensure business continuity. Application disaster tolerance is higher than data disaster tolerance, that is, to establish a complete set of backup application systems that are equivalent to local data systems in different places (you can backup each other with local application systems or work with local application systems). After the disaster, the remote application system quickly took over or assumed the business operation of the local application system.
When designing a disaster recovery backup system, many factors need to be considered, such as the size of backup / restore data, the distance and data transmission method between the application data center and the backup data center, the recovery speed required during the disaster, and the backup Center management and investment funds. According to these factors and different applications, disaster recovery backup can be divided into four levels.
Level 0: No backup center
This level of disaster recovery backup does not actually have disaster recovery capabilities. It only backs up data locally, and the data that is backed up is only saved locally and not sent to another place.
Level 1: Local tape backup, offsite storage
Back up critical data locally and save it offsite. After the disaster, restore the system and data according to the scheduled data recovery procedures. This solution is low cost and easy to configure. However, when the amount of data increases, there is a problem that the storage medium is difficult to manage, and when a disaster occurs, a large amount of data is difficult to recover in time. To solve this problem, when a disaster occurs, recover critical data first, and then recover non-critical data.
Level 2: Hot backup site backup
Establish a hot backup point in a different place, and perform data backup through the network. That is, the data of the main site is backed up to the backup site in a synchronous or asynchronous manner through the network. The backup site generally only backs up data, and does not undertake business. When a disaster occurs, the backup site takes over the business of the primary site, thereby maintaining the continuity of business operations.
Level 3: Activity recovery center
Set up two data centers in far away places, they are in working condition, and carry out mutual data backup. When a disaster occurs in one data center, another data center takes over its work. According to the actual requirements and the amount of capital invested, this level of backup can be divided into two types: â‘ The two data centers are limited to the mutual backup of critical data; Lost etc. Zero data loss is one of the most demanding backup methods for disaster recovery. It requires the system to ensure data security no matter what disaster occurs. Therefore, it requires complex management software and dedicated hardware equipment. The investment is relatively the largest, but the recovery speed is also the fastest.
Key technologies of disaster recovery backup
When building a disaster recovery backup system, various technologies are involved, such as: SAN or NAS technology, remote mirroring technology, IP-based SAN interconnection technology, snapshot technology, etc. Here focuses on remote mirroring, snapshots and interconnection technology.
1. Remote mirroring technology
Remote mirroring technology is used for data backup between the main data center and the backup center. Mirroring is an information storage process that generates a mirrored view of the same data on two or more disks or disk subsystems. One is called the master mirroring system, and the other is called the slave mirroring system. According to the location of the master-slave mirror storage system, it can be divided into local mirror and remote mirror. Remote mirroring, also called remote replication, is the core technology of disaster recovery and backup, and it is also the basis for maintaining remote data synchronization and achieving disaster recovery. Remote mirroring can be divided into synchronous remote mirroring and asynchronous remote mirroring according to whether the host requesting mirroring needs confirmation information from the remote mirroring site.
Synchronous remote mirroring (synchronous replication technology) refers to the use of remote mirroring software to copy local data to remote locations in a completely synchronized manner. Each local I / O transaction needs to wait for the completion confirmation information of remote replication before it is released. Synchronous mirroring allows the copy to always match what the local machine requires to copy. When the primary site fails, after the user's application is switched to the backup alternate site, the mirrored remote copy can ensure that the business continues to execute without data loss. However, it has the disadvantage of long delay caused by round-trip propagation, and is limited to applications at relatively short distances.
Asynchronous remote mirroring (asynchronous replication technology) ensures that the basic operations to the local storage system are completed before the remote storage view is updated, and the local storage system provides confirmation information for the completion of the I / O operation of the requesting mirror host. Remote data replication is performed in a background synchronization manner, which has little impact on the performance of the local system, long transmission distance (up to 1000 kilometers or more), and low requirements on network bandwidth. However, many remote slave storage subsystem writes are not confirmed. When a certain factor causes data transmission to fail, data consistency problems may occur. In order to solve this problem, the technology of delayed replication is mostly used at present (local data replication is performed in the background log area), that is, remote data update is performed after ensuring that the local data is intact.
2. Snapshot technology
Remote mirroring technology is often combined with snapshot technology to achieve remote backup, that is, data is backed up to a remote storage system through mirroring, and then the information in the remote storage system is backed up to a remote tape library or optical disc library using snapshot technology.
Snapshot is to quickly scan the data of the disk subsystem to be backed up by the software to establish a snapshot logical unit number LUN and snapshot cache of the data to be backed up. During the quick scan, the data blocks to be modified during the backup process are quickly copied to the snapshot cache at the same time. The snapshot LUN is a set of pointers that point to the snapshot cache and the unchanged data blocks in the disk subsystem (during the backup process). While the normal business is going on, the snapshot LUN is used to achieve a complete backup of the original data. It enables users to extract current online business data in real time without affecting normal business (mainly referring to disaster recovery backup system). Its "backup window" is close to zero, which can greatly increase the continuity of the system business and provide a guarantee for the real 7 × 24 operation of the system.
Snapshots use memory as a buffer (snapshot cache), and snapshot software provides instant data images of system disk storage. It has buffer scheduling problems.
3. Interconnection technology
Early data backup between the main data center and the backup data center was mainly based on SAN-based remote replication (mirror), that is, the two SANs were connected through Fibre Channel FC to perform remote mirroring (duplication). When a disaster occurs, the backup data center replaces the main data center to ensure the continuity of the system. This remote disaster recovery backup method has some shortcomings, such as: high implementation cost, poor equipment interoperability, and short geographic distance (10 kilometers), etc. These factors hinder its further promotion and application.
Currently, a variety of IP-based SAN remote data disaster recovery technologies have emerged. They use the IP-based SAN interconnection protocol to remotely copy the information in the main data center SAN through the existing TCP / IP network to the backup center SAN. When the amount of data stored in the backup center is too large, you can use the snapshot technology to back it up to the tape library or optical disc library. This remote disaster recovery backup of IP-based SAN can span LAN, MAN and WAN, with low cost, good scalability, and broad development prospects. IP-based interconnection protocols include: FCIP, iFCP, Infiniband, iSCSI, etc.
Phone Wireless Earphones,Sports Bluetooth Earphone,Best Cheap Wireless Earbuds,Wireless Earbuds For Android
Dongguang Vowsound Electronics Co., Ltd. , https://www.vowsound.com