Cross-table inquiry method and related device

Abstract

The invention discloses a cross-table query method, which comprises the following steps: determining index columns in a plurality of tables to be queried, and establishing each index column and corresponding ROWID as index contents in the same index to obtain a cross-table index; when a query instruction is received, the cross-table index is queried according to the query instruction to obtain a corresponding ROWID; selecting a corresponding data acquisition operation according to the number of ROWIDs, and executing the data acquisition operation to obtain a query result. By establishing the same cross-table index for multiple tables, when the tables are queried at the same time, only the cross-table index can be queried to complete the query operation for all contents, thus avoiding the need to perform index query once for each table, reducing multiple queries to one query, reducing the number of queries, and improving the speed of cross-table queries. The invention also discloses a cross-table query system, a server and a computer-readable storage medium, which have the above beneficial effects.

Description

technical field [0001] The present application relates to the field of computer technology, and in particular to a cross-table query method, a cross-table query system, a server, and a computer-readable storage medium. Background technique [0002] In the database, access to the database is usually improved through various index structures, generally including but not limited to B-tree index, bitmap index, and reverse index. Suppose there is a table t1, which has columns (c1, c2, c3), and there are 10,000 records in it (assuming that 100 data blocks are required to store all records). In order to improve the query speed, an index idx1 is created on the column c1, and an index record includes a ROWID (the unique number of each row in the database) and the value of the c1 column, so an index record occupies a small storage space in the database block , a database can contain a large number of index records, so it may only need 10 database blocks to store all the records of th...

AI Technical Summary

Problems solved by technology

However, when the number of records in each index reaches the level of 10,000, when the same condition is queried through two indexes, two separate index query operations will be performed, which usually increases the length of the query.

Furthermore, when the number of tables increases, it will seriously affect the query time of data and reduce query efficiency

Images