Method, device and system for maintaining data consistency and application server

Abstract

The invention discloses a method for maintaining data consistency. The method is executed in an application server and is suitable to be used for simultaneously writing service data into a first database and a second database. The method comprises the following steps of: creating first database connection with a first database, starting a first database transaction under the first database connection, and executing first read-write operation on the first database so as to write service data into the first database; creating second database connection with a second database, starting a second database transaction under the second database connection, and executing second read-write operation on the second database so as to write the service data into the second database; and when a first operation result corresponding to the first read-write operation and a second operation result corresponding to the second read-write operation are both successful, respectively executing the submitting operation on the first database transaction and the second database transaction. The invention also discloses a device and a system for maintaining the data consistency corresponding to the method and the application server.

Description

technical field [0001] The invention relates to the technical field of databases, in particular to a method, device, system and application server for maintaining data consistency. Background technique [0002] Due to the performance and capacity limitations of a single database server, when the business develops to a certain extent, the business data has to be segmented so that the business data is stored in multiple databases of multiple database servers. [0003] When the business needs to write data to multiple databases at the same time, if some databases fail to submit, while other databases submit successfully, there will be inconsistency in business data, resulting in errors in the final stored data that cannot be guaranteed to be consistent. cause loss. [0004] figure 1 It is a flowchart of simultaneous operations on two databases in the prior art. Such as figure 1 As shown, the web server needs to perform cross-database operations and write business data to da...

AI Technical Summary

Problems solved by technology

[0016] It can be seen that according to the execution result of step S109, the operation on database A is successfully executed, but the operation on database B fails, because the operation on database A has been submitted to database A and cannot be rolled back, which will cause the data in database A to be inconsistent with database A. B data is logically inconsistent

Images