Data processing method, device and apparatus for database and index updating method

Abstract

The invention discloses a data processing method for a database. The data processing method comprises: receiving a data insertion request for the database, obtaining a data insertion behavior constraint attribute according to the data insertion request, and inserting data into a main table of the database and an index table associated with the main table according to the data insertion behavior constraint attribute and the data insertion request. By adopting the method, the cost of the data insertion operation for the database is reduced.

Description

technical field [0001] The present application relates to the technical field of databases, in particular to a data processing method, device, electronic equipment and storage equipment for databases. The present application also relates to an index update method. Background technique [0002] In the database field, indexing is an important means to speed up queries, so in addition to the main table of the database, there will also be a database index table. In order to organize the index data, the database system must pay the corresponding operation cost. [0003] First of all, after introducing an index, the original database update operation will be "magnified" to four operations. The data update process of a typical database is as follows: 1. Read the main table to get the original index record; 2. Delete the original index record from the index table; 3. Insert a new record into the index table; 4. Replace the main table The records satisfying the conditions in are u...

AI Technical Summary

Problems solved by technology

The root cause of this problem is: the update operation on the main table row leads to the change of the primary key column of the index table, making it necessary to read the main table once to get the index data to be deleted

[0004] In addition, another scenario of database operation is the insert operation. Since the NoSQL (non-relational database) database system based on the LSM tree structure cannot distinguish whether the database write behavior is an update operation or an insert operation, even during an insert operation, a read operation is required. The operation of the main table of the database is used to determine whether it is an insert operation or an update operation, and the cost of reading the main table operation is relatively large

Images