BitMap索引的效率要优于B-Tree索引实例

一、实验说明：

    操作系统：rhel 5.4 x86

    数据库：Oracle 11g R2

二、操作步骤：

    首先创建一张t_btree表，并建立B-Tree索引，索引键是status：

SQL> create table t_btree as select * from dba_objects;

Table created.

SQL> create index status_btree on t_btree(status);

Index created.

执行两次下面的查询语句，并显示执行计划：

SQL> set autotrace traceonly;
SQL> select count(*) from t_btree where status='VALID';

Execution Plan
----------------------------------------------------------
Plan hash value: 2400455617

--------------------------------------------------------------------------------------
| Id  | Operation          | Name        | Rows  | Bytes | Cost (%CPU)| Time    |
--------------------------------------------------------------------------------------
|  0 | SELECT STATEMENT      |          |      1 |      5 |      49  (0)| 00:00:01 |
|  1 |  SORT AGGREGATE      |          |      1 |      5 |          |        |
|*  2 |  INDEX FAST FULL SCAN| STATUS_BTREE | 74307 |    362K|      49  (0)| 00:00:01 |
--------------------------------------------------------------------------------------

Predicate Information (identified by operation id):
---------------------------------------------------

  2 - filter("STATUS"='VALID')

Note
-----
  - dynamic sampling used for this statement (level=2)

Statistics
----------------------------------------------------------
    32  recursive calls
      0  db block gets
    261  consistent gets
    458  physical reads
      0  redo size
    424  bytes sent via SQL*Net to client
    419  bytes received via SQL*Net from client
      2  SQL*Net roundtrips to/from client
      0  sorts (memory)
      0  sorts (disk)
      1  rows processed

SQL> select count(*) from t_btree where status='VALID';

Execution Plan
----------------------------------------------------------
Plan hash value: 2400455617

--------------------------------------------------------------------------------------
| Id  | Operation          | Name        | Rows  | Bytes | Cost (%CPU)| Time    |
--------------------------------------------------------------------------------------
|  0 | SELECT STATEMENT      |          |      1 |      5 |      49  (0)| 00:00:01 |
|  1 |  SORT AGGREGATE      |          |      1 |      5 |          |        |
|*  2 |  INDEX FAST FULL SCAN| STATUS_BTREE | 74307 |    362K|      49  (0)| 00:00:01 |
--------------------------------------------------------------------------------------

Predicate Information (identified by operation id):
---------------------------------------------------

  2 - filter("STATUS"='VALID')

Note
-----
  - dynamic sampling used for this statement (level=2)

Statistics
----------------------------------------------------------
      0  recursive calls
      0  db block gets
    180  consistent gets
      0  physical reads
      0  redo size
    424  bytes sent via SQL*Net to client
    419  bytes received via SQL*Net from client
      2  SQL*Net roundtrips to/from client
      0  sorts (memory)
      0  sorts (disk)
      1  rows processed

接着创建跟t_btree一样的表t_bmap，并创建BitMap索引。

SQL> create table t_bmap as select * from dba_objects;

Table created.

SQL> create bitmap index status_bmap on t_bmap(status);

Index created.

同样执行之前的语句两次：

SQL> select count(*) from t_bmap where status='VALID';

Execution Plan
----------------------------------------------------------
Plan hash value: 516980546

---------------------------------------------------------------------------------------------
| Id  | Operation              | Name        | Rows  | Bytes | Cost (%CPU)| Time    |
---------------------------------------------------------------------------------------------
|  0 | SELECT STATEMENT          |        |      1 |      5 |      3  (0)| 00:00:01 |
|  1 |  SORT AGGREGATE          |        |      1 |      5 |        |        |
|  2 |  BITMAP CONVERSION COUNT    |        | 62928 |    307K|      3  (0)| 00:00:01 |
|*  3 |    BITMAP INDEX FAST FULL SCAN| STATUS_BMAP |        |        |        |        |
---------------------------------------------------------------------------------------------

Predicate Information (identified by operation id):
---------------------------------------------------

  3 - filter("STATUS"='VALID')

Note
-----
  - dynamic sampling used for this statement (level=2)

Statistics
----------------------------------------------------------
    32  recursive calls
      0  db block gets
    72  consistent gets
    266  physical reads
      0  redo size
    424  bytes sent via SQL*Net to client
    419  bytes received via SQL*Net from client
      2  SQL*Net roundtrips to/from client
      0  sorts (memory)
      0  sorts (disk)
      1  rows processed

SQL> select count(*) from t_bmap where status='VALID';

Execution Plan
----------------------------------------------------------
Plan hash value: 516980546

---------------------------------------------------------------------------------------------
| Id  | Operation              | Name        | Rows  | Bytes | Cost (%CPU)| Time    |
---------------------------------------------------------------------------------------------
|  0 | SELECT STATEMENT          |        |      1 |      5 |      3  (0)| 00:00:01 |
|  1 |  SORT AGGREGATE          |        |      1 |      5 |        |        |
|  2 |  BITMAP CONVERSION COUNT    |        | 62928 |    307K|      3  (0)| 00:00:01 |
|*  3 |    BITMAP INDEX FAST FULL SCAN| STATUS_BMAP |        |        |        |        |
---------------------------------------------------------------------------------------------

Predicate Information (identified by operation id):
---------------------------------------------------

  3 - filter("STATUS"='VALID')

Note
-----
  - dynamic sampling used for this statement (level=2)

Statistics
----------------------------------------------------------
      0  recursive calls
      0  db block gets
      6  consistent gets
      0  physical reads
      0  redo size
    424  bytes sent via SQL*Net to client
    419  bytes received via SQL*Net from client
      2  SQL*Net roundtrips to/from client
      0  sorts (memory)
      0  sorts (disk)
      1  rows processed

从上面的查询中，我们可以得到，分别给两张内容一样的表做查询的时候，在执行第二次的时候是属于软解析：

从一致性读上比较，B-Tree索引的consistent gets是180，BitMap的是6；

从Cost的消耗上看，B-Tree索引的COST是49，而BitMap的是3。

在索引键是高重复率键值(status)的时候情况下BitMap索引的效率要优于B-Tree索引。