<listitem>

<para>

<literal>controlled</literal> &mdash; this mode

uses the default planner for all new queries, but continuescollecting

statistics for already known query types, if any.

uses the default planner for all new queries, but continuesusing

the previously specified planning settings for already known query types, if any.

</para>

</listitem>

<listitem>

</programlisting>

where <replaceable>mode</> is the name of the optimization mode to use.

</para>

<important><para>The intelligent mode of <filename>aqo</filename>

<important><para>The<literal>intelligent</literal> mode of <filename>aqo</filename>

may not work well if the queries in your workload are of multiple

different types. In this case, you can try resetting the optimization mode to <literal>controlled</literal>.</para>

</important>

</programlisting>

<para>

If your data or query distribution is rapidly changing, obsolete statistics may affect <filename>aqo</filename> performance. After a while, <filename>aqo</filename> will learn the new statistics, but it may take longer than expected. To speed up learning, reset the

If your data or query distribution is rapidly changing, learning new statistics will take longer than usual.

In this case, obsolete statistics may affect performance. To speed up <filename>aqo</filename> learning, reset the

statistics. To remove all the collected machine learning

statistics, run the following command:

</para>