Embedded-database-system-oriented no journaling transaction processing method

Abstract

The method provides an embedded-database-system-oriented no journaling transaction processing method. An embedded database comprises two work areas which are used for maintaining data in the database and recording the operation of a user on the database. The two work areas manage an identical data object and comprise identical quantity of object elements. Elements in the each work area are positioned through indexing. In an initial state at the beginning of a transaction, the object elements in the two work areas corresponding to an identical index value point to an identical data object, in a submission state or returning state of the transaction, object elements in the two work areas are compared, if the object elements in the two work areas are inconsistent, the data object is proved to be modified, and then the validation or the canceling of the modification is determined, and the whole object elements in the work areas are processed by the same method. The transaction processing mechanism provided by the invention has advantages of powerful functions, high efficiency and fewer occupied resources.

Description

technical field [0001] The invention relates to a database system transaction processing method, in particular to a log-free transaction processing method oriented to an embedded database system, and belongs to the technical field of database management. Background technique [0002] With the development of technology, embedded devices and mobile devices have been widely used in various industries, such as: handheld PDA, industrial computer, smart phone, etc. And almost all embedded devices and mobile devices will involve data management. It is impossible to run a large general-purpose database on embedded devices and mobile devices, so how to achieve stable and efficient data management has become a difficult problem for embedded application developers. [0003] Embedded database has been widely used in embedded devices because of its easy-to-use, small size, and strong data management capabilities. However, most of the current embedded databases all or partly follow the ...

AI Technical Summary

Problems solved by technology

Access to the disk during transaction execution will inevitably affect the performance of the database, and saving log files on the disk also increases the storage overhead of the database on the embedded device

This kind of storage overhead can be basically ignored on PCs or servers, but on embedded devices with slow processors and tight disk resources, the increased disk overhead caused by running a database system that needs to record logs is often unbearable

Especially in embedded applications with high performance requirements, embedded databases that need to record logs are even more inapplicable

Images