Searching more partitions than necessary

Question

I have a fact table that is partitioned by month:
PARTITION BY RANGE (request_timestamp)
( PARTITION requests_2004_01 VALUES LESS THAN (TO_DATE('2004-02', 'YYYY-MM'))
PARTITION requests_2004_02 VALUES LESS THAN (TO_DATE('2004-03', 'YYYY-MM'))
PARTITION requests_xxxx_xx VALUES LESS THAN (MAXVALUE) );
Now if I do
SELECT COUNT(*) FROM requests
WHERE
time >= TO_DATE('2005-02','YYYY-MM') AND
time < TO_DATE('2005-03','YYYY-MM');
the query executes fast, and the plan contains "PARTITION RANGE (SINGLE)", as expected.
But if I do
SELECT COUNT(*) FROM requests
WHERE
time >= TRUNC(ADD_MONTHS(CURRENT_DATE, -1), 'MONTH') AND
time < TRUNC(CURRENT_DATE, 'MONTH');
the query runs more than ten times slower, with "PARTITION RANGE (ITERATOR)".
Do I really have to hardcode the date literals, or is there a way to make the optimizer aware of the fact that it needs to look into one partition only (but without hardcoding the partition)?

Answer 1

actually your first query is according to optimizer range or in other words optimizer can understand it well and your second query using two function to get actual date so you should try here function based indexes on the column but for good performance you should use optimize queries like first one. avoide those queries which misdirect the optimizer