# SQL Statements Panel Columns Filter
+ Monitor the Impact of SQL Statements on TiKV Performance
  + Mean Visible Versions Per Query
  + Mean Meet Versions Per Query
![`smart`](TiDBMonitorSQLPerf03_en_001_20240125.png)
------------------------------------------------------------------------------
When using the `SQL statements` feature, we need to pay attention to the system's memory usage and can configure it through `settings` to prevent excessive memory usage. First, we can set the number of stored statements through 'Max # Statement.' When the number of stored statements exceeds the configuration of 'Max # Statement,' old statements will be discarded. Secondly, we can divide time into time windows with a specified size in minutes, and the maximum history data that we can display is the window count multiplied by the window size. For example, in the above figure, if the window size is 5 minutes and the window count is 4, the maximum historical data that we can store is 4 x 5 (minutes) = 20 (minutes), so any SQL statements older than 20 minutes will be discarded. 
From the above configuration, we can see that the larger 'Max # Statement' or the larger the (`Number of Window` x `Window Size`), the more historical SQL statements we can collect, but this also consumes more memory. Therefore, we need to configure it carefully to avoid memory overflow in the SQL statement analysis feature.

3. The amount of data read from the storage layer of each SQL statement execution, multiplied by the cumulative number of SQL statement executions, allowing us to find the SQL statement that reads the most data.