Continuous data protection method based on centralized storage

Abstract

The invention provides a continuous data protection method based on centralized storage; difference algorithm is adopted to carry on difference transmission of files in data transmission process, the file changes are recorded through the combination of mirror and difference for storage, the system operates in Linux operating system, the user protected data and the metadata information are directly stored in the file system of the server, the system creates a metadata directory, a storage protection instance list and user information content for each user, also includes a soft link pointing at an physical data storage space, the time data storage space allocates an independent directory, logic volume and a partition for different user groups, and subsequently creates correspondingly directories for the users, a directory as the subdirectory of the user directory is created for each protection instance for protecting all versions of difference data of the protection instance; the invention is an important component part of project chunk level and file level continuous data protection system and helps obtain good benefits in part of enterprise key data protection services.

Description

technical field [0001] The invention relates to a file-based continuous data protection system, which can capture the change of a single file in real time and provide file recovery at any point in time. The differential algorithm used in data transmission performs differential transmission of files, and the storage uses a combination of mirror image and differential to record file changes. The method effectively utilizes network bandwidth and saves storage resources. Background technique [0002] For an enterprise, data security is extremely important. Once important data is damaged or lost, it will have a major impact on the daily production of the enterprise, and even cause irreparable losses. Most enterprises use traditional data backup technologies for data protection, such as backup, RAID, remote mirroring, and snapshots. These methods have their own purposes, and the data protection of large enterprises generally needs to integrate these methods. However, tradition...

AI Technical Summary

Problems solved by technology

However, traditional data protection solutions focus on the periodic backup of data, so there are always problems such as backup window and impact on the production system, and it is difficult to guarantee flexible target recovery point (Recovery point objectives, RPO) and faster Recovery time objectives (RTO), and RPO and RTO have become key indicators to measure the emergency response capability and data protection capability of the disaster recovery system

Images