查询重写是指当对物化视图的基表进行查询时，Oracle会自动判断能否通过查询物化视图来得到结果，如果可以，则避免了聚集或连接操作，而直接从已经计算好的物化视图中读取数据。
1. 在scott下做测试,一个普通的2表关联查询
> alter system flush buffer_cache;
收集统计信息
> exec dbms_stats.gather_table_stats('SCOTT','EMP',cascade=>true);
> exec dbms_stats.gather_table_stats('SCOTT','DEPT',cascade=>true);
> set autotrace traceonly
> select ename,job,dname from emp,dept where emp.deptno=dept.deptno;
14 rows selected.
PLAN_TABLE_OUTPUT
Plan hash value: 844388907
----------------------------------------------------------------------------------------
| Id | Operation | Name | Rows | Bytes | Cost (%CPU)| Time |
----------------------------------------------------------------------------------------
| 0 | SELECT STATEMENT | | 14 | 420 | 6 (17)| 00:00:01 |
| 1 | MERGE JOIN | | 14 | 420 | 6 (17)| 00:00:01 |
| 2 | TABLE ACCESS BY INDEX ROWID | DEPT | 4 | 52 | 2 (0)| 00:00:01 |
| 3 | INDEX FULL SCAN | PK_DEPT | 4 | | 1 (0)| 00:00:01 |
|* 4 | SORT JOIN | | 14 | 238 | 4 (25)| 00:00:01 |
| 5 | TABLE ACCESS FU | EMP | 14 | 238 | 3 (0)| 00:00:01 |
----------------------------------------------------------------------------------------
Statistics
----------------------------------------------------------
1 recursive calls
0 db block gets
10 consistent gets
8 physical reads
0 redo size
970 bytes sent via SQL*Net to client
519 bytes received via SQL*Net from client
2 SQL*Net roundtrips to/from client
1 sorts (memory)
0 sorts (disk)
14 rows processed
以上，物理读和一致性读分别是8，10
2. 创建以上对象的物化视图，并查看通过物化视图查询所得到的执行计划以及统计信息
CREATE MATERIALIZED VIEW LOG ON emp
WITH ROWID
INCLUDING NEW VALUES;
CREATE MATERIALIZED VIEW LOG ON dept
WITH ROWID
INCLUDING NEW VALUES;
CREATE MATERIALIZED VIEW emp_dept_mv_01
BUILD IMMEDIATE
REFRESH force ON COMMIT
as select ename,job,dname
from emp,dept
where emp.deptno=dept.deptno;
> alter system flush buffer_cache;
收集统计信息，可以针对物化视图
> exec dbms_stats.gather_table_stats('SCOTT','EMP_DEPT_MV_01',cascade=>true);
> select * from emp_dept_mv_01;
Execution Plan
----------------------------------------------------------
Plan hash value: 2905768236
---------------------------------------------------------------------------------------
| Id | Operation
| Name | Rows | Bytes | Cost (%CPU)| Time |
---------------------------------------------------------------------------------------
| 0 | SELECT STATEMENT
| | 14 | 308 | 3 (0)| 00:00:01 |
| 1 | MAT_VIEW ACCESS FULL | EMP_DEPT_MV_01 | 14 | 308 | 3
(0)| 00:00:01 |
---------------------------------------------------------------------------------------
Statistics
----------------------------------------------------------
0 recursive calls
0 db block gets
4 consistent gets
2 physical reads
0 redo size
1079 bytes sent via SQL*Net to client
519 bytes received via SQL*Net from client
2 SQL*Net roundtrips to/from client
0 sorts (memory)
0 sorts (disk)
14 rows processed
通过物化视图的查询，发现只有4个一致性读和2个物理读（相比先前的10，8），说明物化视图会为系统带来一定的性能提升
3. 将物化视图设置为enable query rewrite(默认为disable),并执行原查询语句，查看执行计划
> alter MATERIALIZED VIEW EMP_DEPT_MV_01 enable query rewrite;
> alter system flush buffer_cache;
> select ename,job,dname from emp,dept where emp.deptno=dept.deptno;
14 rows selected.
Execution Plan
----------------------------------------------------------
Plan hash value: 1320338361
-----------------------------------------------------------------------------------------------
| Id | Operation
| Name | Rows | Bytes | Cost (%CPU)| Time |
-----------------------------------------------------------------------------------------------
| 0 | SELECT STATEMENT
| | 14 | 308 | 3 (0)| 00:00:01 |
| 1 | MAT_VIEW REWRITE ACCESS FULL| EMP_DEPT_MV_01 | 14 | 308 | 3
(0)| 00:00:01 |
-----------------------------------------------------------------------------------------------
Statistics
----------------------------------------------------------
1 recursive calls
0 db block gets
3 consistent gets
2 physical reads
0 redo size
1079 bytes sent via SQL*Net to client
519 bytes received via SQL*Net from client
2 SQL*Net roundtrips to/from client
0 sorts (memory)
0 sorts (disk)
14 rows processed
该语句实际上自动查询的是与其对应的物化视图。
一致性读/物理读由原来的(10,8)，下降到了(3,2),效果明显.
总结：
特定场景中，利用物化视图可提高查询速度，但需要额外的存储空间；
查询重写: 无需改变代码，只需创建一个支持查询重写的物化视图即可提高检索性能