Oracle 10.2.0.4 Index on timestamp column not used when SYSTIMESTAMP Used.

Question

Hi,
I have a table A with a column B timestamp(6). The tables contains around 300000 rows..
I have created index 'idx' on the column B.
When i compare column 'B' with systimestamp, it does not use the index, whereas if i compare 'B' with sysdate it uses the index.
Eg :
select count(*) from a where b<=sysdate;
The above used the index 'idx' and executed in 1 second
select count(*) from a where b<=systimestamp;
The above does not use the index and executed in 19 seconds.
Any Clue?
Thanks in Advance

Answer 1

Oracle is using Internal functions when you use SYSTIMESTAMP:
Work around will be to use TO_TIMESTAMP as shown below.. Or define a function based index..
You can check performance problem querying against a "TIMESTAMP WITH TIME ZONE" column link also
SQL> create table a(b timestamp(6));
Table created.
SQL> insert into a
  2  select systimestamp+(level/24) from dual connect by level <= 30000;
30000 rows created.
SQL> commit;
Commit complete.
SQL> create index ndx on a(b);
Index created.
SQL> explain plan for
  2  select count(*) from a where b <= sysdate;
Explained.
SQL> select * from table(dbms_xplan.display);
PLAN_TABLE_OUTPUT
Plan hash value: 3858831102
| Id  | Operation         | Name | Rows  | Bytes | Cost (%CPU)| Time     |
|   0 | SELECT STATEMENT  |      |     1 |    13 |     1   (0)| 00:00:01 |
|   1 |  SORT AGGREGATE   |      |     1 |    13 |            |          |
|*  2 |   INDEX RANGE SCAN| NDX  |     1 |    13 |     1   (0)| 00:00:01 |
Predicate Information (identified by operation id):
   2 - access("B"<=SYSDATE@!)
Note
   - dynamic sampling used for this statement
18 rows selected.
SQL> explain plan for
  2  select count(*) from a where b <= systimestamp;
Explained.
SQL>  select * from table(dbms_xplan.display);
PLAN_TABLE_OUTPUT
Plan hash value: 3918351354
| Id  | Operation          | Name | Rows  | Bytes | Cost (%CPU)| Time     |
|   0 | SELECT STATEMENT   |      |     1 |    13 |    20  (15)| 00:00:01 |
|   1 |  SORT AGGREGATE    |      |     1 |    13 |            |          |
|*  2 |   TABLE ACCESS FULL| A    |     1 |    13 |    20  (15)| 00:00:01 |
Predicate Information (identified by operation id):
   2 - filter(SYS_EXTRACT_UTC(INTERNAL_FUNCTION("B"))<=SYS_EXTRACT_UTC(S
              YSTIMESTAMP(6)))
Note
   - dynamic sampling used for this statement
19 rows selected.
--"Just tried using TO_TIMESTAMP"
SQL> explain plan for
  2  select count(*) from a where b <= to_timestamp(systimestamp);
Explained.
SQL>
SQL> select * from table(dbms_xplan.display);
PLAN_TABLE_OUTPUT
Plan hash value: 3858831102
| Id  | Operation         | Name | Rows  | Bytes | Cost (%CPU)| Time     |
|   0 | SELECT STATEMENT  |      |     1 |    13 |     2   (0)| 00:00:01 |
|   1 |  SORT AGGREGATE   |      |     1 |    13 |            |          |
|*  2 |   INDEX RANGE SCAN| NDX  |     4 |    52 |     2   (0)| 00:00:01 |
Predicate Information (identified by operation id):
   2 - access("B"<=TO_TIMESTAMP(TO_CHAR(SYSTIMESTAMP(6))))
14 rows selected.Edited by: jeneesh on Oct 23, 2012 11:18 AM