关系型数据库的管理方法及装置

By periodically traversing the first sub-data table in memory and loading data from persistent storage, the problem of low read performance in relational databases based on LSM Trees is solved, thereby improving data query efficiency and ensuring data correctness.

CN115794960BActive Publication Date: 2026-07-17BEIJING OCEANBASE TECHNOLOGY CO LTD
View PDF 2 Cites 0 Cited by

Patent Information

Authority / Receiving Office
CN · China
Patent Type
Patents(China)
Current Assignee / Owner
BEIJING OCEANBASE TECHNOLOGY CO LTD
Filing Date
2022-11-01
Publication Date
2026-07-17

AI Technical Summary

Technical Problem

Existing relational databases based on LSM Trees have low read performance during data queries and cannot be effectively optimized.

Method used

Periodically traverse the first sub-data table in memory, identify candidate data rows whose access count exceeds a threshold, and load unstored data columns from the second sub-data table in persistent storage into the first sub-data table to ensure the integrity of the data stored in memory.

Benefits of technology

It improves data query efficiency, avoids disk access, and ensures the timeliness and accuracy of data.

✦ Generated by Eureka AI based on patent content.

Smart Images

  • Figure CN115794960B_ABST
    Figure CN115794960B_ABST
Patent Text Reader

Abstract

本申请一个或多个实施例提供一种关系型数据库的管理方法及装置；关系型数据库为基于LSM Tree的关系型数据库；关系型数据库中的目标数据表包括第一子数据表和第二子数据表；第一子数据表为缓存在内存中的数据表；第二子数据表为存储在用于持久化存储数据的存储介质中的数据表；该方法包括：按照预设的第一时长周期，定期遍历第一子数据表，并确定第一子数据表是否包括访问次数大于预设的第一阈值的候选数据行；如果是，则确定第一子数据表中是否存储了候选数据行中与目标数据表中的数据列对应的全列数据；如果否，则从第二子数据表中读取候选数据行中与第一子数据表中未存储的数据列对应的数据，并将读取到的数据加载到第一子数据表中。
Need to check novelty before this filing date? Find Prior Art