Supercharge Your Innovation With Domain-Expert AI Agents!

Multi-site disaster recovery backup method based on distributed raid slice

A disaster recovery backup, distributed technology, applied in response to the generation of errors, redundancy in operations for data error detection, input/output to record carriers, etc.

Pending Publication Date: 2021-04-16
军事科学院系统工程研究院网络信息研究所
View PDF0 Cites 0 Cited by
  • Summary
  • Abstract
  • Description
  • Claims
  • Application Information

AI Technical Summary

Problems solved by technology

[0003] At present, the commonly used two-site three-center disaster recovery backup solution is based on remote disk mirroring. However, the single-site data in the disaster recovery center of this solution is complete. When a disaster occurs in the disaster recovery center, the backup data is safe and secure. There is an urgent need to establish a networked disaster recovery backup method with high reliability and high security due to the risk of reliability

Method used

the structure of the environmentally friendly knitted fabric provided by the present invention; figure 2 Flow chart of the yarn wrapping machine for environmentally friendly knitted fabrics and storage devices; image 3 Is the parameter map of the yarn covering machine
View more

Image

Smart Image Click on the blue labels to locate them in the text.
Viewing Examples
Smart Image
  • Multi-site disaster recovery backup method based on distributed raid slice
  • Multi-site disaster recovery backup method based on distributed raid slice
  • Multi-site disaster recovery backup method based on distributed raid slice

Examples

Experimental program
Comparison scheme
Effect test

Embodiment 1

[0049] Embodiment 1: Data backup

[0050] S101, the data backup operation is carried out sequentially according to the sequence of events, for example, the three data backup events of backup data 0, backup data 1 and backup data 2 are carried out in sequence according to the order of 0, 1, and 2;

[0051] S102. When there is new data to be backed up, the production center divides the data to be backed up into data strips of a fixed size, such as image 3 Backup data 0 is divided into three stripes: stripe 0, stripe 1, and stripe 2;

[0052] S103. According to the disaster recovery topology and distributed RAID strategy, perform slice and check calculation on the data stripe, and send the check data and the source data slice to the disaster recovery site; image 3 Among them, 10 disaster recovery sites (disaster recovery sites 0-9) and the RAID strategy of RAID5+PQ verification are used to slice the data stripe into 8 data blocks D1-D8, and two verification data of P and Q are...

Embodiment 2

[0055] Example 2: Data Recovery

[0056] When the data in the production center is damaged, the data needs to be restored to the recovery target center (it can be the original production center or another production center). The recovery steps are as follows.

[0057] S201. Data restoration in the production center takes stripes as a unit and restores one by one;

[0058] S202. The recovery target center reads the source data slice data block of stripe 0 in the disaster recovery site through the Internet, such as Figure 4 As shown, read disaster recovery site 0-7 when stripe 0 recovers;

[0059] S203. The recovery target center reorganizes the data according to the original data sequence (D1-D8) according to the RAID strategy;

[0060] S204. Save the reorganized data to the recovery target center storage, and complete the data recovery of stripe 0;

[0061] S205. Steps S202 to S204 are repeated according to stripes until data of all stripes is restored, that is, data resto...

Embodiment 3

[0062] Example 3: Data reconstruction

[0063] If PQ verification is adopted, data reconstruction and data recovery can be supported when the data of at most two disaster recovery sites is damaged at the same time. Such as Figure 5 As shown, assuming that disaster recovery sites 8 and 9 are damaged, the data reconstruction process is as follows.

[0064] S301. The process of data reconstruction, taking data stripes as a unit, restores data stripe by stripe;

[0065] S302. The production center reads the 8 data blocks stored in the disaster recovery site 0-7 in the stripe 0 through the Internet;

[0066] S303. The production center calculates the data blocks of the disaster recovery site 8 and the disaster recovery site 9 from the read 8 data blocks according to the RAID strategy and the PQ verification algorithm;

[0067] S304. The production center sends the calculated data blocks to the disaster recovery site 8 and the disaster recovery site 9 respectively through the In...

the structure of the environmentally friendly knitted fabric provided by the present invention; figure 2 Flow chart of the yarn wrapping machine for environmentally friendly knitted fabrics and storage devices; image 3 Is the parameter map of the yarn covering machine
Login to View More

PUM

No PUM Login to View More

Abstract

The invention discloses a multi-site disaster recovery backup method based on distributed RAID slices, and belongs to the technical field of data disaster recovery storage. According to the method, source data of a production center is subjected to data slicing and verification calculation according to a specific raid strategy, and sliced data and verification data are dispersedly backed up to a plurality of disaster recovery sites which are distributed in a wide area and interconnected with a network. The method has the characteristics that firstly, the survivability of the system is improved, and data can be recovered through other disaster recovery stations on the network after a part of disaster recovery stations are integrally damaged; and 2, the data content secrecy capability is improved, any disaster recovery site cannot obtain effective data, and the effective data can be obtained only by intercepting and capturing data of a certain number of disaster recovery sites. According to the invention, data disaster recovery backup service with invulnerability, obstinate storage, security, secrecy and wide-area distribution can be provided for an information system.

Description

technical field [0001] The invention belongs to the technical field of data disaster recovery backup, and specifically discloses a multi-site disaster recovery backup method based on distributed raid slicing. Background technique [0002] With the continuous development of information technology, the amount of data generated by application systems deployed in data centers has increased rapidly, reaching TB or even PB levels. Data reliability is the basis for the reliable operation of information systems. Current data protection technologies mainly include snapshots, cloning, remote file replication, and remote disk mirroring. Among them, snapshots and clones store static data at a certain point in time. When the data source changes, the previously stored snapshots and clones will not change, so the purpose of real-time data backup cannot be achieved. Remote file replication is a file-level backup solution. The files to be backed up are transmitted to remote disaster recover...

Claims

the structure of the environmentally friendly knitted fabric provided by the present invention; figure 2 Flow chart of the yarn wrapping machine for environmentally friendly knitted fabrics and storage devices; image 3 Is the parameter map of the yarn covering machine
Login to View More

Application Information

Patent Timeline
no application Login to View More
IPC IPC(8): G06F3/06G06F11/14
Inventor 张凯谢永强宁希刘志波李忠博
Owner 军事科学院系统工程研究院网络信息研究所
Features
  • R&D
  • Intellectual Property
  • Life Sciences
  • Materials
  • Tech Scout
Why Patsnap Eureka
  • Unparalleled Data Quality
  • Higher Quality Content
  • 60% Fewer Hallucinations
Social media
Patsnap Eureka Blog
Learn More