Association rule-based database file page prefetching method and device

Abstract

The embodiment of the invention provides a database file page prefetching method and device based on association rules. The method comprises the steps that a currently read first file page and the number requirement of prefetched file pages are obtained; based on the first file page, determining a corresponding second file page in the disk according to an association rule between preset file pages; determining a storage mode of the second file page based on a relationship between the number requirement of the prefetched file pages and a preset storage threshold value; based on the storage mode, third file pages meeting the requirement for the number of the pre-fetched file pages are extracted from the second file pages and loaded to the memory. Therefore, the access trend of the data can be reflected by utilizing a plurality of association rules existing in the database, the file pages are prefetched through the association rules, the database prefetching accuracy can be improved, the database waiting time can be shortened, the database response speed can be improved, and the data access efficiency can be improved by flexibly determining the storage mode of the prefetched file pages in the memory. The maximum database performance improvement is achieved by consuming the minimum memory, CPU and disk resources.

Description

technical field [0001] The invention relates to the technical field of databases, in particular to a method and device for prefetching database file pages based on association rules. Background technique [0002] In a database system, a file page (page for short) is the smallest unit of disk storage and management. The process of database accessing data after receiving sql request is: [0003] 1. Calculate the file page of the data in the disk. [0004] 2. Load the file page from disk to memory. [0005] 3. Read data from memory. [0006] In this process, the time consumption of calculating the data location and reading data from memory is negligible, and the most time-consuming step is loading data from disk to memory. In the process of processing SQL requests, the database can process memory data and disk IO in parallel; if you can read the file page (page) that needs to be loaded in the next step while processing memory data, you can reduce waiting for disk IO time, ...

AI Technical Summary

Problems solved by technology

[0008] In view of this, the embodiment of the present invention provides a method and device for prefetching database file pages based on association rules, so as to overcome the linear prefetching method of file pages based on the principle of program locality in the prior art. The consumption of disk and disk resources is large, which affects the response speed of the database

Images