# Adding an Index to the `Probe Side`

![`smart`](TuningJoinByExplainAnalyze03_en_001_20230129.png)

------------------------------------------------------------------------------
In the slide, the `actRows` value for the `IndexHashJoin_14` operator matches the `actRows` value of its child operator, `IndexLookUp_11(Probe)`, both being 892 rows. Here, the `EMPLOYEES` table serves as the driving table, the `SALARIES` table as the driven table, and an index exists on the join column of the driven table. This indicates that the current execution plan is optimal.

This example demonstrates that when analyzing SQL queries involving table joins, the process involves first addressing filtering issues for each table and then selecting the most efficient join algorithm. Both steps can be effectively evaluated using the `EXPLAIN ANALYZE` output.