10g和11g，优化器对外连接的处理对比

我反省，今天面试有个问题没有说清楚。
我给出的结论（而且这个结论我验证过）是：不要使用不必要的外连接，举了下面这个例子却没有说清楚。虽然最近感冒，状态不是很好，但最擅长的东西都没有表达清楚，泪流满面啊：（
多谢那位面试官DBA的提醒，我以后要多多注意啊，要把自己擅长的东西说清楚。

10g：
test@ORA10G> select * from v$version;

BANNER
----------------------------------------------------------------
Oracle Database 10g Enterprise Edition Release 10.2.0.5.0 - 64bi
PL/SQL Release 10.2.0.5.0 - Production
CORE    10.2.0.5.0      Production
TNS for IBM/AIX RISC System/6000: Version 10.2.0.5.0 - Productio
NLSRTL Version 10.2.0.5.0 - Production

创建测试环境：
drop table t1;
drop table t2;

create table t1 as select * from dba_objects;
create table t2 as select * from t1;

create index i_t2_objectid on t2(object_id) parallel 3;
alter index i_t2_objectid noparallel;

analyze table t1 compute statistics for table for all indexes for all indexed columns;
analyze table t2 compute statistics for table for all indexes for all indexed columns;

执行SQL并查看其执行计划：
test@ORA10G> select t1.object_id, t2.object_id, t1.object_name, t2.object_name
  2    from t1, t2
  3   where t1.object_id = t2.object_id(+)
  4     and t2.object_id in (3, 6, 11, 28);

Execution Plan
----------------------------------------------------------
Plan hash value: 1594007018

---------------------------------------------------------------------------------------
| Id  | Operation              | Name | Rows  | Bytes |TempSpc| Cost (%CPU)| Time     |
---------------------------------------------------------------------------------------
|   0 | SELECT STATEMENT       |      |     4 |   600 |       | 32331   (1)| 00:06:28 |
|*  1 |  FILTER                |      |       |       |       |            |          |
|*  2 |   HASH JOIN RIGHT OUTER|      |     4 |   600 |   173M| 32331   (1)| 00:06:28 |
|   3 |    TABLE ACCESS FULL   | T2   |  2196K|   148M|       |  7109   (1)| 00:01:26 |
|   4 |    TABLE ACCESS FULL   | T1   |  2196K|   165M|       |  7110   (1)| 00:01:26 |
---------------------------------------------------------------------------------------

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

   1 - filter("T2"."OBJECT_ID"=3 OR "T2"."OBJECT_ID"=6 OR "T2"."OBJECT_ID"=11
              OR "T2"."OBJECT_ID"=28)
   2 - access("T1"."OBJECT_ID"="T2"."OBJECT_ID"(+))

Statistics
----------------------------------------------------------
          1  recursive calls
          0  db block gets
      64216  consistent gets
          0  physical reads
          0  redo size
        855  bytes sent via SQL*Net to client
        492  bytes received via SQL*Net from client
          2  SQL*Net roundtrips to/from client
          0  sorts (memory)
          0  sorts (disk)
          4  rows processed

从上面结果可以看到，即使t2中字段object_id上有索引，在10g中也是用不到的。Oracle先选择将两个表走hash join right outer连接，然后将结果进行filter过滤。

同样的操作，下面看11g将会怎样处理：
SQL> select * from v$version;

BANNER
--------------------------------------------------------------------------------
Oracle Database 11g Enterprise Edition Release 11.2.0.1.0 - 64bit Production
PL/SQL Release 11.2.0.1.0 - Production
CORE    11.2.0.1.0      Production
TNS for Linux: Version 11.2.0.1.0 - Production
NLSRTL Version 11.2.0.1.0 - Production

创建测试环境：
drop table t1;
drop table t2;

create table t1 as select * from dba_objects;
create table t2 as select * from t1;

create index i_t2_objectid on t2(object_id) parallel 3;
alter index i_t2_objectid noparallel;

analyze table t1 compute statistics for table for all indexes for all indexed columns;
analyze table t2 compute statistics for table for all indexes for all indexed columns;

执行SQL并查看其执行计划：
SQL> set linesize 400 pagesize 400
SQL> set autot trace
SQL> 
SQL> select t1.object_id,t1.object_name,t2.object_id,t2.object_name
  2  from t1,t2
  3  where t1.object_id=t2.object_id(+)
  4  and t2.object_id in (3,6,11,28);

Execution Plan
----------------------------------------------------------

Plan hash value: 1372208711

-----------------------------------------------------------------------------------------------
| Id  | Operation                     | Name          | Rows  | Bytes | Cost (%CPU)| Time     |
-----------------------------------------------------------------------------------------------
|   0 | SELECT STATEMENT              |               |    12 |  1896 |    57   (2)| 00:00:01 |
|*  1 |  HASH JOIN                    |               |    12 |  1896 |    57   (2)| 00:00:01 |
|   2 |   INLIST ITERATOR             |               |       |       |            |          |
|   3 |    TABLE ACCESS BY INDEX ROWID| T2            |    12 |   948 |     5   (0)| 00:00:01 |
|*  4 |     INDEX RANGE SCAN          | I_T2_OBJECTID |     4 |       |     4   (0)| 00:00:01 |
|*  5 |   TABLE ACCESS FULL           | T1            |   131 | 10349 |    51   (0)| 00:00:01 |
-----------------------------------------------------------------------------------------------

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

   1 - access("T1"."OBJECT_ID"="T2"."OBJECT_ID")
   4 - access("T2"."OBJECT_ID"=3 OR "T2"."OBJECT_ID"=6 OR "T2"."OBJECT_ID"=11 OR
              "T2"."OBJECT_ID"=28)
   5 - filter("T1"."OBJECT_ID"=3 OR "T1"."OBJECT_ID"=6 OR "T1"."OBJECT_ID"=11 OR
              "T1"."OBJECT_ID"=28)

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

从上面可以看出，11g中，Oracle是忽略了外连接的。Oracle在11g中优化器应该做了优化。
我今天本来想表达的意思是：10g中外连接在这种情况下其实是先两个表full table scan的结果做了hash 外连接，将hash的结果做fillter过滤。而11g会忽略掉外连接，将t2的id列会走索引，然后跟另一个表做join连接。