Chinaunix首页 | 论坛 | 博客
  • 博客访问: 8472
  • 博文数量: 7
  • 博客积分: 0
  • 博客等级: 民兵
  • 技术积分: 10
  • 用 户 组: 普通用户
  • 注册时间: 2015-06-04 14:50
文章分类
文章存档

2016年(7)

我的朋友
最近访客

分类: Oracle

2016-01-23 21:22:16

      因为优化器还不够强大,还有很多限制,或者因为一些逻辑原因,分析认为SQL要走索引比较好,但是事实却无法正确利用索引。这时候,除了给ORACLE需要的统计信息之外,写的SQL必须要能够给优化器足够多的额外有效信息,让优化器能够选择更好的执行计划。
要让给优化器正确使用上需要的索引,要考虑两点:
1).如何避免优化器的限制 2).根据业务数据特点改写SQL语句

      说明:这里说的走不了索引,是指走不了正常的RANGE SCAN,非(FAST) FULL INDEX SCAN。

1.谓词使用不等于(<>,!=),目前的优化器一般认为不等于的选择性不好,会找到很多行,所以用不上索引,就算强制索引,也是扫描全部的索引,走不了RANGE SCAN
   如果事实走索引定位较好,那么解决方法:
   1)如果不等条件之外的值不多,而且是确认的,可以改为等值或IN查询,比如status状态字段一般值类别很少
   
   2)如果不等条件之外的值很多,可以改为> OR <的形式,当然第2种方法包含了方法1.

 例如:
  DROP TABLE  t;
 CREATE TABLE t AS SELECT LEVEL ID,rpad('x',10,'x') padding,CAST('JACK' AS VARCHAR2(100)) NAME
 FROM dual
 CONNECT BY LEVEL<100000;
--构造3行较少的值

 INSERT INTO t VALUES(100000,'y','DINGJUN1');
 INSERT INTO t VALUES(100001,'y','DINGJUN2');
 INSERT INTO t VALUES(100002,'y','DINGJUN3');
 COMMIT;
 ALTER TABLE t MODIFY NAME NOT NULL;
--创建索引

 CREATE INDEX idx_t ON t(NAME);
 BEGIN
  dbms_stats.gather_table_stats(ownname => USER,tabname => 'T',estimate_percent => 100,method_opt => 'for columns name size   10',cascade => TRUE);
 END;
 /



<>无法利用索引:
dingjun123@ORADB>  SELECT * FROM t
  2   WHERE t.name<>'JACK';
3 rows selected.
Elapsed: 00:00:00.01
Execution Plan
----------------------------------------------------------
Plan hash value: 1601196873
--------------------------------------------------------------------------
| Id  | Operation         | Name | Rows  | Bytes | Cost (%CPU)| Time     |
--------------------------------------------------------------------------
|   0 | SELECT STATEMENT  |      |     4 |    88 |   107   (2)| 00:00:02 |
|*  1 |  TABLE ACCESS FULL| T    |     4 |    88 |   107   (2)| 00:00:02 |
--------------------------------------------------------------------------
Predicate Information (identified by operation id):
---------------------------------------------------
   1 - filter("T"."NAME"<>'JACK')

将<>改写为OR连接后,能够正确使用索引,走OR扩展:
dingjun123@ORADB>  SELECT * FROM t
  2   WHERE t.NAME>'JACK' OR t.NAME <'JACK';
3 rows selected.
Elapsed: 00:00:00.06
Execution Plan
----------------------------------------------------------
Plan hash value: 2945726203
--------------------------------------------------------------------------------------
| Id  | Operation                    | Name  | Rows  | Bytes | Cost (%CPU)| Time     |
--------------------------------------------------------------------------------------
|   0 | SELECT STATEMENT             |       |     4 |    88 |     6   (0)| 00:00:01 |
|   1 |  CONCATENATION               |       |       |       |            |          |
|   2 |   TABLE ACCESS BY INDEX ROWID| T     |     1 |    22 |     3   (0)| 00:00:01 |
|*  3 |    INDEX RANGE SCAN          | IDX_T |     1 |       |     2   (0)| 00:00:01 |
|   4 |   TABLE ACCESS BY INDEX ROWID| T     |     3 |    66 |     3   (0)| 00:00:01 |
|*  5 |    INDEX RANGE SCAN          | IDX_T |     4 |       |     2   (0)| 00:00:01 |
--------------------------------------------------------------------------------------
Predicate Information (identified by operation id):
---------------------------------------------------
   3 - access("T"."NAME">'JACK')
   5 - access("T"."NAME"<'JACK')
       filter(LNNVL("T"."NAME">'JACK'))

如果业务允许,改为写下列语句也是走索引的,不再演示。
 
  SELECT * FROM t 
 WHERE t.NAME IN ('DINGJUN1','DINGJUN2','DINGJUN3');
 
 NOT IN,NOT EXISTS与<>类似,也是走不了索引的(注意,这里说走不了索引,只针对于不等连接谓词,不包括其他谓词)
 演示NOT IN:
 DROP TABLE t1;
 CREATE TABLE t1 AS SELECT 1 ID,'JACK' NAME FROM dual;
 ALTER TABLE t1 MODIFY NAME NOT NULL;
  BEGIN
  dbms_stats.gather_table_stats(ownname => USER,tabname => 'T1',estimate_percent => 100);
 END;
 /


   NOT IN也不能使用索引,强制使用只能扫描全部索引:
dingjun123@ORADB>   SELECT * FROM t
  2    WHERE t.NAME NOT IN(SELECT  t1.NAME FROM t1);
Elapsed: 00:00:00.00
Execution Plan
----------------------------------------------------------
Plan hash value: 4253491563
-----------------------------------------------------------------------------
| Id  | Operation            | Name | Rows  | Bytes | Cost (%CPU)| Time     |
-----------------------------------------------------------------------------
|   0 | SELECT STATEMENT     |      |   100K|  2636K|   110   (2)| 00:00:02 |
|*  1 |  HASH JOIN RIGHT ANTI|      |   100K|  2636K|   110   (2)| 00:00:02 |
|   2 |   TABLE ACCESS FULL  | T1   |     1 |     5 |     3   (0)| 00:00:01 |
|   3 |   TABLE ACCESS FULL  | T    |   100K|  2148K|   106   (1)| 00:00:02 |
-----------------------------------------------------------------------------
Predicate Information (identified by operation id):
---------------------------------------------------
   1 - access("T"."NAME"="T1"."NAME")


dingjun123@ORADB>   SELECT/*+index(t idx_t)*/ * FROM t
  2    WHERE t.NAME NOT IN(SELECT  t1.NAME FROM t1);
Elapsed: 00:00:00.03
Execution Plan
----------------------------------------------------------
Plan hash value: 1707532828
--------------------------------------------------------------------------------------
| Id  | Operation                    | Name  | Rows  | Bytes | Cost (%CPU)| Time     |
--------------------------------------------------------------------------------------
|   0 | SELECT STATEMENT             |       |   100K|  2636K|   591   (1)| 00:00:08 |
|   1 |  MERGE JOIN ANTI             |       |   100K|  2636K|   591   (1)| 00:00:08 |
|   2 |   TABLE ACCESS BY INDEX ROWID| T     |   100K|  2148K|   587   (1)| 00:00:08 |
|   3 |    INDEX FULL SCAN           | IDX_T |   100K|       |   225   (1)| 00:00:03 |
|*  4 |   SORT UNIQUE                |       |     1 |     5 |     4  (25)| 00:00:01 |
|   5 |    TABLE ACCESS FULL         | T1    |     1 |     5 |     3   (0)| 00:00:01 |
--------------------------------------------------------------------------------------
Predicate Information (identified by operation id):
---------------------------------------------------
   4 - access("T"."NAME"="T1"."NAME")
       filter("T"."NAME"="T1"."NAME")

  NOT EXISTS与NOT IN类似(注意,NOT IN已经处理过NULL的影响
   SELECT * FROM t 
  WHERE  NOT EXISTS(SELECT 1  FROM t1 WHERE t.NAME=t1.NAME);

2.由于B*Tree索引不存储全为NULL的行,造成无法走索引的情况
 解决方法:
   1)视业务要求是否可以不统计NULL的
   加条件where xxx IS NOT NULL, COUNT(列).....等等过滤NULL的语句
   
   2)复合列索引
   
   3)函数索引
 详细可以参考【Oracle index】B*Tree索引之NULL统计技巧 这个帖子里的内容。

3.LIKE前通配查询询,LIKE全通配查询走不了索引
解决方法:
  1)是否可以根据业务需求把前通配去掉
dingjun123@ORADB> SELECT * FROM t WHERE t.NAME LIKE '%DINGJUN%';
Elapsed: 00:00:00.04
Execution Plan
----------------------------------------------------------
Plan hash value: 1601196873
--------------------------------------------------------------------------
| Id  | Operation         | Name | Rows  | Bytes | Cost (%CPU)| Time     |
--------------------------------------------------------------------------
|   0 | SELECT STATEMENT  |      |  5000 |   107K|   107   (2)| 00:00:02 |
|*  1 |  TABLE ACCESS FULL| T    |  5000 |   107K|   107   (2)| 00:00:02 |
--------------------------------------------------------------------------
Predicate Information (identified by operation id):
---------------------------------------------------
   1 - filter("T"."NAME" LIKE '%DINGJUN%')
--改为后通配,走索引
dingjun123@ORADB> SELECT * FROM t WHERE t.NAME LIKE 'DINGJUN%';
Elapsed: 00:00:00.00
Execution Plan
----------------------------------------------------------
Plan hash value: 1594971208
-------------------------------------------------------------------------------------
| Id  | Operation                   | Name  | Rows  | Bytes | Cost (%CPU)| Time     |
-------------------------------------------------------------------------------------
|   0 | SELECT STATEMENT            |       |     3 |    66 |     3   (0)| 00:00:01 |
|   1 |  TABLE ACCESS BY INDEX ROWID| T     |     3 |    66 |     3   (0)| 00:00:01 |
|*  2 |   INDEX RANGE SCAN          | IDX_T |     3 |       |     2   (0)| 00:00:01 |
-------------------------------------------------------------------------------------
Predicate Information (identified by operation id):
---------------------------------------------------
   2 - access("T"."NAME" LIKE 'DINGJUN%')
       filter("T"."NAME" LIKE 'DINGJUN%')


  2) 是否和此LIKE一样的前通配或全通配的SQL有很多,此谓词的LIKE变化不大,如果是,考虑建立函数索引,否则对于全通配问题最好办法就是全文索引

dingjun123@ORADB>   CREATE INDEX idx1_t ON t (instr(NAME,'DINGJUN'));
Index created.
Elapsed: 00:00:00.16dingjun123@ORADB>  SELECT * FROM t WHERE instr(t.NAME,'DINGJUN')>0;

Elapsed: 00:00:00.00
Execution Plan
----------------------------------------------------------
Plan hash value: 2071967826
--------------------------------------------------------------------------------------
| Id  | Operation                   | Name   | Rows  | Bytes | Cost (%CPU)| Time     |
--------------------------------------------------------------------------------------
|   0 | SELECT STATEMENT            |        |  5000 |   102K|     7   (0)| 00:00:01 |
|   1 |  TABLE ACCESS BY INDEX ROWID| T      |  5000 |   102K|     7   (0)| 00:00:01 |
|*  2 |   INDEX RANGE SCAN          | IDX1_T |   900 |       |     3   (0)| 00:00:01 |
--------------------------------------------------------------------------------------
Predicate Information (identified by operation id):
---------------------------------------------------
   2 - access(INSTR("NAME",'DINGJUN')>0)

  3)如果只是前通配,可以使用reverse函数索引(不是翻转键索引)
  SELECT  * FROM t WHERE t.NAME LIKE '%DINGJUN1';
  CREATE INDEX idx2_t ON t(REVERSE(NAME));
  语句要改写,使用reverse函数谓词,并且查找值倒查:
dingjun123@ORADB> SELECT * FROM t WHERE REVERSE(t.NAME) LIKE '1NUJGNID%';
Elapsed: 00:00:00.00
Execution Plan
----------------------------------------------------------
Plan hash value: 3787301248
--------------------------------------------------------------------------------------
| Id  | Operation                   | Name   | Rows  | Bytes | Cost (%CPU)| Time     |
--------------------------------------------------------------------------------------
|   0 | SELECT STATEMENT            |        |  5000 |   102K|     8   (0)| 00:00:01 |
|   1 |  TABLE ACCESS BY INDEX ROWID| T      |  5000 |   102K|     8   (0)| 00:00:01 |
|*  2 |   INDEX RANGE SCAN          | IDX2_T |   900 |       |     4   (0)| 00:00:01 |
--------------------------------------------------------------------------------------
Predicate Information (identified by operation id):
---------------------------------------------------
   2 - access(REVERSE("NAME") LIKE '1NUJGNID%')
       filter(REVERSE("NAME") LIKE '1NUJGNID%')

注意:如果查找的是中文,得注意,不可直接写 REVERSE(t.NAME) LIKE '1NUJGNID%',因为REVERSE内部会按字节翻转,中文的写法可以用REVERSE转换,比如
SELECT  * FROM t WHERE REVERSE(t.name) LIKE REVERSE('数据')||'%';否则查询出的数据不对。


4.对索引列使用了函数,数学运算,其他表达式等
 解决方法:去掉对索引列的相关运算,保持索引列纯净。
 优化器目前对一些数学运算,还无法做很好的消除动作,所以对于索引列应该尽量保持纯净,否则可能无法用上正确的索引
dingjun123@ORADB>  CREATE INDEX idx3_t ON t(ID);
Index created.
Elapsed: 00:00:00.13

dingjun123@ORADB>  ALTER TABLE T MODIFY ID NOT NULL;
Table altered.
Elapsed: 00:00:00.10

dingjun123@ORADB> SELECT * FROM t WHERE ID+0=1;
Elapsed: 00:00:00.00
Execution Plan
----------------------------------------------------------
Plan hash value: 1601196873
--------------------------------------------------------------------------
| Id  | Operation         | Name | Rows  | Bytes | Cost (%CPU)| Time     |
--------------------------------------------------------------------------
|   0 | SELECT STATEMENT  |      |  1000 | 21000 |   107   (2)| 00:00:02 |
|*  1 |  TABLE ACCESS FULL| T    |  1000 | 21000 |   107   (2)| 00:00:02 |
--------------------------------------------------------------------------
Predicate Information (identified by operation id):
---------------------------------------------------
   1 - filter("ID"+0=1)

dingjun123@ORADB> SELECT * FROM t WHERE ID=1;
Elapsed: 00:00:00.01
Execution Plan
----------------------------------------------------------
Plan hash value: 2351669764
--------------------------------------------------------------------------------------
| Id  | Operation                   | Name   | Rows  | Bytes | Cost (%CPU)| Time     |
--------------------------------------------------------------------------------------
|   0 | SELECT STATEMENT            |        |  1000 | 21000 |     3   (0)| 00:00:01 |
|   1 |  TABLE ACCESS BY INDEX ROWID| T      |  1000 | 21000 |     3   (0)| 00:00:01 |
|*  2 |   INDEX RANGE SCAN          | IDX3_T |   400 |       |     1   (0)| 00:00:01 |
--------------------------------------------------------------------------------------
Predicate Information (identified by operation id):
---------------------------------------------------
 2 - access("ID"=1)

  举例只是举简单的数学运算,可能运算还和其他列运算,比如where ID+ext_col...

5.ORACLE使用了隐式类型转换,导致索引无法使用
  解决方法:

  必须避免隐式类型转换,全部要求显式类型转换(非索引列),且避免对索引列进行类型转换(有函数索引除外)
  如果类型不一致,不管是否发生自动类型转换,谓词的右值应该显式转换为与索引列保持一致(对于非索引列的运算也应该如此)。

dingjun123@ORADB>  DROP TABLE t1;
Table dropped.
Elapsed: 00:00:00.03

dingjun123@ORADB>  CREATE TABLE t1(x VARCHAR2(100));
Table created.
Elapsed: 00:00:00.04

dingjun123@ORADB>  CREATE INDEX idx_t1 ON t1(x);
Index created.
Elapsed: 00:00:00.04

dingjun123@ORADB>  SELECT * FROM t1 WHERE x = 1;
Elapsed: 00:00:00.00
Execution Plan
----------------------------------------------------------
Plan hash value: 3617692013
--------------------------------------------------------------------------
| Id  | Operation         | Name | Rows  | Bytes | Cost (%CPU)| Time     |
--------------------------------------------------------------------------
|   0 | SELECT STATEMENT  |      |     1 |    52 |     2   (0)| 00:00:01 |
|*  1 |  TABLE ACCESS FULL| T1   |     1 |    52 |     2   (0)| 00:00:01 |
--------------------------------------------------------------------------
Predicate Information (identified by operation id):
---------------------------------------------------
   1 - filter(TO_NUMBER("X")=1)

dingjun123@ORADB>  SELECT * FROM t1 WHERE x = '1';
Elapsed: 00:00:00.00
Execution Plan
----------------------------------------------------------
Plan hash value: 1369807930
---------------------------------------------------------------------------
| Id  | Operation        | Name   | Rows  | Bytes | Cost (%CPU)| Time     |
---------------------------------------------------------------------------
|   0 | SELECT STATEMENT |        |     1 |    52 |     1   (0)| 00:00:01 |
|*  1 |  INDEX RANGE SCAN| IDX_T1 |     1 |    52 |     1   (0)| 00:00:01 |
---------------------------------------------------------------------------
Predicate Information (identified by operation id):
---------------------------------------------------
   1 - access("X"='1')


 发生类型转换,可以通过执行计划中的谓词信息获知。通过分析发现,X因为是VARCHAR2,优先级比数值类型低,遇到数值类型,会TO_NUMBER,所以索引失效,通过传入与索引列类型一致的字符串后,解决。

6.对于其他不走索引的,尽量通过业务的逻辑特点改写语句,可能正确利用索引
 例如一个典型的例子:
DROP TABLE t_objects;
CREATE TABLE t_objects
AS
SELECT * FROM dba_objects;
CREATE INDEX idx_t_objects ON t_objects(last_ddl_time,created);
 BEGIN
  dbms_stats.gather_table_stats(ownname => USER,tabname => 't_objects',estimate_percent => 100,cascade => TRUE);
 END;
/

需求:查找创建时间是2013年的,并且最后ddl时间比创建时间大1天以上。
dingjun123@ORADB> SELECT * FROM
  2  t_objects t
  3  WHERE t.last_ddl_time-t.created>1
  4  AND t.created>=DATE'2013-1-1';
Elapsed: 00:00:00.00
Execution Plan
----------------------------------------------------------
Plan hash value: 3629755566
-------------------------------------------------------------------------------
| Id  | Operation         | Name      | Rows  | Bytes | Cost (%CPU)| Time     |
-------------------------------------------------------------------------------
|   0 | SELECT STATEMENT  |           |  1171 |   110K|   300   (1)| 00:00:04 |
|*  1 |  TABLE ACCESS FULL| T_OBJECTS |  1171 |   110K|   300   (1)| 00:00:04 |
-------------------------------------------------------------------------------
Predicate Information (identified by operation id):
---------------------------------------------------
   1 - filter("T"."CREATED">=TO_DATE(' 2013-01-01 00:00:00',
              'syyyy-mm-dd hh24:mi:ss') AND "T"."LAST_DDL_TIME"-"T"."CREATED">1)

  

    这个索引是组合索引,上面的语句对前导列进行了运行,也不符合走index skip scan的条件,所以,走FULL TABLE SCAN。那么是否可以通过逻辑改写走索引呢,通过第4点得知,要保持索引列清净,将create_date移到右边,语句如下
 
dingjun123@ORADB> SELECT   * FROM
  2  t_objects t
  3  WHERE  t.last_ddl_time>=(t.created+1)
  4  AND t.created>=DATE'2013-1-1';
Elapsed: 00:00:00.00

Execution Plan
----------------------------------------------------------
Plan hash value: 3629755566
-------------------------------------------------------------------------------
| Id  | Operation         | Name      | Rows  | Bytes | Cost (%CPU)| Time     |
-------------------------------------------------------------------------------
|   0 | SELECT STATEMENT  |           |   403 | 39091 |   301   (1)| 00:00:04 |
|*  1 |  TABLE ACCESS FULL| T_OBJECTS |   403 | 39091 |   301   (1)| 00:00:04 |
-------------------------------------------------------------------------------
Predicate Information (identified by operation id):
---------------------------------------------------
   1 - filter("T"."CREATED">=TO_DATE(' 2013-01-01 00:00:00',
              'syyyy-mm-dd hh24:mi:ss') AND "T"."LAST_DDL_TIME">=INTERNAL_FUNCTION("T"
              ."CREATED")+1)

    通过改写后发现,还是没有走索引,因为ORACLE认为前导列右边的created不固定,无法从指定索引处查找。通过这个分析得知,ORACLE谓词传递有一定限制,create_date+1无法做谓词传递给last_ddl_time,改写:
dingjun123@ORADB> SELECT   * FROM
  2  t_objects t
  3  WHERE t.last_ddl_time>=(t.created)
  4  AND t.last_ddl_time>=(t.created+1)
  5  AND t.created>=DATE'2013-1-1';
Elapsed: 00:00:00.00

Execution Plan
----------------------------------------------------------
Plan hash value: 641904483
---------------------------------------------------------------------------------------------
| Id  | Operation                   | Name          | Rows  | Bytes | Cost (%CPU)| Time     |
---------------------------------------------------------------------------------------------
|   0 | SELECT STATEMENT            |               |     2 |   194 |    11   (0)| 00:00:01 |
|   1 |  TABLE ACCESS BY INDEX ROWID| T_OBJECTS     |     2 |   194 |    11   (0)| 00:00:01 |
|*  2 |   INDEX RANGE SCAN          | IDX_T_OBJECTS |     6 |       |    10   (0)| 00:00:01 |
---------------------------------------------------------------------------------------------
Predicate Information (identified by operation id):
---------------------------------------------------
   2 - access("T"."LAST_DDL_TIME">=TO_DATE(' 2013-01-01 00:00:00', 'syyyy-mm-dd
              hh24:mi:ss') AND "T"."CREATED">=TO_DATE(' 2013-01-01 00:00:00', 'syyyy-mm-dd
              hh24:mi:ss') AND "T"."LAST_DDL_TIME" IS NOT NULL)
       filter("T"."CREATED">=TO_DATE(' 2013-01-01 00:00:00', 'syyyy-mm-dd
              hh24:mi:ss') AND "T"."LAST_DDL_TIME">="T"."CREATED" AND
              "T"."LAST_DDL_TIME">=INTERNAL_FUNCTION("T"."CREATED")+1)

   上面的oracle知道谓词传递给last_ddl_time了,T"."LAST_DDL_TIME">=TO_DATE(' 2013-01-01 00:00:00', 'syyyy-mm-dd hh24:mi:ss'),也可以手动谓词传递,last_ddl_time肯定大于等于DATE'2013-1-2'
dingjun123@ORADB> SELECT   * FROM
  2  t_objects t
  3  WHERE t.last_ddl_time>=DATE'2013-1-2'
  4  AND t.last_ddl_time>=(t.created+1)
  5  AND t.created>=DATE'2013-1-1';
Elapsed: 00:00:00.00
Execution Plan
----------------------------------------------------------
Plan hash value: 641904483
---------------------------------------------------------------------------------------------
| Id  | Operation                   | Name          | Rows  | Bytes | Cost (%CPU)| Time     |
---------------------------------------------------------------------------------------------
|   0 | SELECT STATEMENT            |               |    13 |  1261 |    11   (0)| 00:00:01 |
|   1 |  TABLE ACCESS BY INDEX ROWID| T_OBJECTS     |    13 |  1261 |    11   (0)| 00:00:01 |
|*  2 |   INDEX RANGE SCAN          | IDX_T_OBJECTS |    13 |       |    10   (0)| 00:00:01 |
---------------------------------------------------------------------------------------------
Predicate Information (identified by operation id):
---------------------------------------------------
   2 - access("T"."LAST_DDL_TIME">=TO_DATE(' 2013-01-02 00:00:00', 'syyyy-mm-dd
              hh24:mi:ss') AND "T"."CREATED">=TO_DATE(' 2013-01-01 00:00:00', 'syyyy-mm-dd
              hh24:mi:ss') AND "T"."LAST_DDL_TIME" IS NOT NULL)
       filter("T"."CREATED">=TO_DATE(' 2013-01-01 00:00:00', 'syyyy-mm-dd
              hh24:mi:ss') AND "T"."LAST_DDL_TIME">=INTERNAL_FUNCTION("T"."CREATED")+1)

   如果查询条件无t.created>=DATE'2013-1-2',如下面语句:
SELECT * FROM
    t_objects t
    WHERE t.last_ddl_time-t.created>1;

QL> set autotrace traceonly exp
QL>  SELECT * FROM
 2      t_objects t
 3      WHERE t.last_ddl_time-t.created>1;

执行计划
---------------------------------------------------------
Plan hash value: 3629755566
------------------------------------------------------------------------------
 Id  | Operation         | Name      | Rows  | Bytes | Cost (%CPU)| Time     |
------------------------------------------------------------------------------
   0 | SELECT STATEMENT  |           | 11706 |  1108K|   297   (2)| 00:00:04 |
*  1 |  TABLE ACCESS FULL| T_OBJECTS | 11706 |  1108K|   297   (2)| 00:00:04 |
------------------------------------------------------------------------------


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


  1 - filter("T"."LAST_DDL_TIME"-"T"."CREATED">1)

SQL> set autotrace off
SQL>  SELECT COUNT(*) FROM
  2      t_objects t;


  COUNT(*)
----------
     74059


SQL>  SELECT COUNT(*) FROM
  2      t_objects t
  3      WHERE t.last_ddl_time-t.created>1;


  COUNT(*)
----------
       216

应该走索引更佳,因为没有其他过滤条件,可以考虑建立函数索引:
SQL>     CREATE INDEX idx1_t_object ON t_objects(last_ddl_time-created);
索引已创建。

--注意收集直方图,因为分布不均
SQL> exec dbms_stats.gather_table_stats(ownname => USER,tabname => 't_objects',estimate_percent => 100,method_opt => 'for all indexed columns',cascade => TRUE);
PL/SQL 过程已成功完成。

SQL>      SELECT COUNT(*) FROM
  2      t_objects t
  3      WHERE t.last_ddl_time-t.created>1;
执行计划
----------------------------------------------------------
Plan hash value: 3236535878
-----------------------------------------------------------------------------------
| Id  | Operation         | Name          | Rows  | Bytes | Cost (%CPU)| Time     |
-----------------------------------------------------------------------------------
|   0 | SELECT STATEMENT  |               |     1 |     9 |     4   (0)| 00:00:01 |
|   1 |  SORT AGGREGATE   |               |     1 |     9 |            |          |
|*  2 |   INDEX RANGE SCAN| IDX1_T_OBJECT |   986 |  8874 |     4   (0)| 00:00:01 |
-----------------------------------------------------------------------------------
Predicate Information (identified by operation id):
---------------------------------------------------
   2 - access("LAST_DDL_TIME"-"CREATED">1)
统计信息
----------------------------------------------------------
          1  recursive calls
          0  db block gets
          3  consistent gets          
         0  physical reads

          0  redo size
        424  bytes sent via SQL*Net to client
        416  bytes received via SQL*Net from client
          2  SQL*Net roundtrips to/from client
          0  sorts (memory)
          0  sorts (disk)
          1  rows processed

   当然,对于两个都是范围的查询,这里只能通过一个列来轮询索引,先做access,再做filter。SQL语句的逻辑改写很重要,往往通过逻辑改写就能改变SQL的执行计划,从不好的计划到好的计划,比如semi join,anti join与or,往往走FILTER导致执行计划较差,这时候就需要通过逻辑等价改写,逻辑等价改写往往需要掌握一些集合的知识,比如NOT (A AND B)===NOT A OR NOT B,NOT (A OR B)===NOT A AND NOT B等。
  
阅读(315) | 评论(0) | 转发(0) |
给主人留下些什么吧!~~
评论热议
请登录后评论。

登录 注册