一、引言 DBA在日常维护管理数据库进行低性能SQL分析时，有时候需要通过创建索引对SQL进行优化，但有些时候我们创建的索引是否能用到？这个只能创建以后才能看出效果，但是在实际工作中，特别是对大表创建索引对系统性能有很大影响，因此我们不得不避开业务高峰时段，但是有没有一种办法创建索引而不影响性能呢？有，那就是虚拟索引。 虚拟索引不是物理存在的，它并不会创建实际的索引段，只是在数据字典中加了一个索引的记录，使得优化器能够意识到一个索引的存在，从而判断是否使用该索引作为访问路径。作用仅仅是为了DBA作SQL优化时使用，DBA根据虚拟索引的优化效果决定是否创建物理索引。 二、虚拟索引类型 虚拟索引支持B-TREE索引和BIT位图索引，在CBO模式下ORACLE优化器会考虑虚拟索引，但是在RBO模式下需要添加hint才行。 三、虚拟索引创建实例 SQL> create table scott.t as select * from dba_objects; Table created. SQL> alter session set "_use_nosegment_indexes"=true; Session altered. SQL> create index scott.ix_t_id on scott.t(object_id) nosegment; Index created. SQL> set autot traceonly SQL> select * from scott.t where object_id=1; no rows selected Execution Plan ---------------------------------------------------------- Plan hash value: 206018885 --------------------------------------------------------------------------------------- | Id | Operation | Name | Rows | Bytes | Cost (%CPU)| Time | --------------------------------------------------------------------------------------- | 0 | SELECT STATEMENT | | 14 | 2898 | 5 (0)| 00:00:01 | | 1 | TABLE ACCESS BY INDEX ROWID| T | 14 | 2898 | 5 (0)| 00:00:01 | |* 2 | INDEX RANGE SCAN | IX_T_ID | 330 | | 1 (0)| 00:00:01 | --------------------------------------------------------------------------------------- Predicate Information (identified by operation id): --------------------------------------------------- 2 - access("OBJECT_ID"=1) Note ----- - dynamic sampling used for this statement (level=2) Statistics ---------------------------------------------------------- 4 recursive calls 0 db block gets 1308 consistent gets 1239 physical reads 0 redo size 1343 bytes sent via SQL*Net to client 512 bytes received via SQL*Net from client 1 SQL*Net roundtrips to/from client 0 sorts (memory) 0 sorts (disk) 0 rows processed -- 以下看的是真实执行计划，显然是用不到索引。 SQL> set autot off SQL> alter session set statistics_level=all; Session altered. SQL> select * from scott.t where object_id=1; no rows selected SQL> select * from table(dbms_xplan.display_cursor(null,null,'allstats last')); PLAN_TABLE_OUTPUT ------------------------------------------------------------------------------------- SQL_ID 2qhwh0nzrzx2r, child number 1 ------------------------------------- select * from t where object_id=1 Plan hash value: 1601196873 --------------------------------------------------------------------------------------------- | Id | Operation | Name | Starts | E-Rows | A-Rows | A-Time | Buffers | Reads | --------------------------------------------------------------------------------------------- | 0 | SELECT STATEMENT | | 1 | | 0 |00:00:00.52 | 1242 | 1239 | |* 1 | TABLE ACCESS FULL| T | 1 | 14 | 0 |00:00:00.52 | 1242 | 1239 | --------------------------------------------------------------------------------------------- Predicate Information (identified by operation id): --------------------------------------------------- 1 - filter("OBJECT_ID"=1) Note ----- - dynamic sampling used for this statement (level=2) 22 rows selected. -- 从数据字段中是无法找到这个索引的。 SQL> select index_name,status from dba_indexes where table_name='T'; no rows selected
四、虚拟索引的特点
4.1、虚拟索引无法执行alter index选项
SQL> alter index scott.IX_T_ID rebuild;
alter index scott.IX_T_ID rebuild
*
ERROR at line 1:
ORA-08114: can not alter a fake index
4.2、使用回收站特性的时候，虚拟索引必须先drop，才能创建同名的索引。
SQL> drop table scott.t;
Table dropped.
SQL> flashback table scott.t to before drop;
Flashback complete.
SQL> create index scott.idx_t_id on scott.t(object_id) nosegment;
create index scott.idx_t_id on scott.t(object_id) nosegment
*
ERROR at line 1:
ORA-00955: name is already used by an existing object
4.3、不能创建和虚拟索引同名的实际索引；
4.4、可以创建和虚拟索引包含相同列但不同名的实际索引;
4.5、虚拟索引在数据字典里看不到
