Updated on 2024-11-05 GMT+08:00

Case: Adjusting the Table Storage Mode in a Medium Table

In GaussDB(DWS), row-store tables use the row execution engine, and column-store tables use the column execution engine. If both row-store table and column-store tables exist in a SQL statement, the system will automatically select the row execution engine. The performance of a column execution engine (except for the indexscan related operators) is much better than that of a row execution engine. Therefore, a column-store table is recommended. This is important for some medium result set dumping tables, and you need to select a proper table storage type.

Before Optimization

During the test at a site, if the following execution plan is performed, the customer expects that the performance can be improved and the result can be returned within 3s.

After Optimization

It is found that the row engine is used after analysis, because both the temporary plan table input_acct_id_tbl and the medium result dumping table row_unlogged_table use a row-store table.

After the two tables are changed into column-store tables, the system performance is improved and the result is returned by 1.6s.