test6
PL/SQL procedure successfully completed.
Elapsed: 00:03:35.33
SQL> select count(*) from t;
COUNT(*)
----------
2000000
Elapsed: 00:00:00.08
Execution Plan
----------------------------------------------------------
0 SELECT STATEMENT Optimizer=ALL_ROWS (Cost=811 Card=1)
1 0 SORT (AGGREGATE)
2 1 TABLE ACCESS (FULL) OF 'T' (TABLE) (Cost=811 Card=1767807)
SQL> create index id_ind on t( id); <--在Id列上建一个索引
Index created.
Elapsed: 00:00:26.74
SQL> select * from t where id = '200'; <---用字符串查，可以使用到索引（请看执行计划）。
ID
--------------------
200
Elapsed: 00:00:00.01 <---根据索引，只用了00.01秒
Execution Plan
----------------------------------------------------------
0 SELECT STATEMENT Optimizer=ALL_ROWS (Cost=3 Card=1 Bytes=12)
1 0 INDEX (RANGE SCAN) OF 'ID_IND' (INDEX) (Cost=3 Card=1 Bytes=12)
从执行计划中可以看出，cost只要3
SQL> select * from t where id = 200; <----如果直接根据数字查，由于发生了隐式转换，执行计划为全表扫描。
ID
--------------------
200
Elapsed: 00:00:00.48 <---全表扫描，是索引扫描时间的48倍
Execution Plan
----------------------------------------------------------
0 SELECT STATEMENT Optimizer=ALL_ROWS (Cost=881 Card=38 Bytes= 456)
1 0 TABLE ACCESS (FULL) OF 'T' (TABLE) (Cost=881 Card=38 Bytes=456)