Redundancy backup method suitable for wide-area distributed real-time database

Abstract

The present invention discloses a redundancy backup method suitable for a wide-area distributed real-time database. In a wide-area system, one or more copies are established in a distributed real-time database system for data in the system in a redundant manner, and consistency of the system is ensured by management of the copies. The method comprises the following steps: (1) the copies are divided into a cold copy and a thermal copy, wherein the thermal copy provides an external access, and the cold copy is only used as a backup copy providing no external service capability, and a thermal copy as a benchmark in all thermal copies becomes a master copy, and only the master copy is allowed to be written; (2) managing a status of the copy, wherein the copy is managed by means of status conversion the copy such as master/standby data switching and cold/thermal copy switching, so as to ensure consistency of data of the system; and (3) performing secondary positioning on a request that the master copy cannot achieve. According to the redundancy backup method suitable for a wide-area distributed real-time database provided by the present invention, a redundancy mechanism between different servers and different systems in the system is implemented; and consistency of the data in the wide-area distributed system is ensured by management of the copies.

Description

[0001] Field [0002] The invention relates to a redundant backup method suitable for wide-area distributed real-time databases, belonging to the field of wide-area distributed real-time database systems. Background technique [0003] The wide-area distributed real-time database system is composed of real-time database systems distributed on several systems. For the scalability of each subsystem and to realize the mutual backup mechanism between subsystems, the subsystems are built on different servers inside the subsystem. Redundant nodes within the system, while constructing a redundancy mechanism for subsystems between subsystems, but the existing real-time database system does not implement a redundancy mechanism between different servers in the system and between systems, so it is difficult to ensure that the wide-area distributed system internal data consistency. Contents of the invention [0004] In order to overcome the deficiencies in the prior art, the purpose...

AI Technical Summary

Problems solved by technology

[0003] The wide-area distributed real-time database system is composed of real-time database systems distributed on several systems. For the scalability of each subsystem and to realize the mutual backup mechanism between subsystems, the subsystems are built on different servers inside the subsystem. Redundant nodes within the system, while constructing a redundancy mechanism for subsystems between subsystems, but the existing real-time database system does not implement a redundancy mechanism between different servers in the system and between systems, so it is difficult to ensure that the wide-area distributed system internal data consistency

Images