Question about selectivity estimation for InnoDB and ha_innodb::records_in_range -
05-16-2006
, 08:32 PM
------=_Part_21283_21840573.1147829535700
Content-Type: text/plain; charset=ISO-8859-1; format=flowed
Content-Transfer-Encoding: quoted-printable
Content-Disposition: inline
ha_innodb::records_in_range uses btr_estimate_n_rows_in_range to estimate
the selectivity of a predicate for an index. btr_estimate_n_rows_in_range
scans the root block of the index to estimate the value.
Are these estimates less accurate immediately after a split of the root
block? There will be few keys in the root block after it has been split.
Mark Callaghan
mcallaghan (AT) google (DOT) com
------=_Part_21283_21840573.1147829535700-- |