還是一樣，

16:45:44 SQL> l

1??select??site_id,

2? ?? ?? ?count(*) sendnum

3??from

4??(select site_id,

5? ?? ?? ?ewb_no

6? ? from prod_send t

7? ? where scan_time >= to_date('2015-05-15', 'yyyy-mm-dd')

8? ???and scan_time < to_date('2015-05-16', 'yyyy-mm-dd')

9? ? group by site_id, ewb_no)

10* group by site_id

16:45:51 SQL> /

2793 rows selected.

Elapsed: 00:00:03.71

Execution Plan

----------------------------------------------------------

Plan hash value: 652676349

--------------------------------------------------------------------------------------------

| Id??| Operation? ?? ?? ?? ?| Name? ?? ???| Rows??| Bytes |TempSpc| Cost (%CPU)| Time? ???|

--------------------------------------------------------------------------------------------

|? ?0 | SELECT STATEMENT? ???|? ?? ?? ?? ? |??3995 | 15980 |? ?? ? |? ?162K??(1)| 00:32:36 |

|? ?1 |??HASH GROUP BY? ?? ? |? ?? ?? ?? ? |??3995 | 15980 |? ?? ? |? ?162K??(1)| 00:32:36 |

|? ?2 |? ?VIEW? ?? ?? ?? ?? ?| VM_NWVW_0? ?|? ?437K|??1708K|? ?? ? |? ?162K??(1)| 00:32:36 |

|? ?3 |? ? HASH GROUP BY? ???|? ?? ?? ?? ? |? ?437K|? ? 10M|? ? 15M|? ?162K??(1)| 00:32:36 |

|*??4 |? ???TABLE ACCESS FULL| PROD_SEND??|? ?437K|? ? 10M|? ?? ? |? ?159K??(1)| 00:31:59 |

--------------------------------------------------------------------------------------------

Predicate Information (identified by operation id):

---------------------------------------------------

4 - filter("SCAN_TIME">=TO_DATE(' 2015-05-15 00:00:00', 'syyyy-mm-dd

hh24:mi:ss') AND "SCAN_TIME"

hh24:mi:ss'))

Statistics

----------------------------------------------------------

1??recursive calls

0??db block gets

586772??consistent gets

586762??physical reads

0??redo size

57840??bytes sent via SQL*Net to client

2570??bytes received via SQL*Net from client

188??SQL*Net roundtrips to/from client

0??sorts (memory)

0??sorts (disk)

2793??rows processed

單獨執行里面的子句，也是走全表

16:48:21 SQL> l

1??select site_id,

2? ?? ?ewb_no

3??from prod_send t

4? ?where scan_time >= to_date('2015-05-15', 'yyyy-mm-dd')

5? ?and scan_time < to_date('2015-05-16', 'yyyy-mm-dd')

6*??group by site_id, ewb_no

16:48:24 SQL> /

850824 rows selected.

Elapsed: 00:00:08.87

Execution Plan

----------------------------------------------------------

Plan hash value: 2203583913

------------------------------------------------------------------------------------------

| Id??| Operation? ?? ?? ?? ?| Name? ?? ???| Rows??| Bytes |TempSpc| Cost (%CPU)| Time? ? |

------------------------------------------------------------------------------------------

|? ?0 | SELECT STATEMENT? ???|? ?? ?? ?? ? |? ?437K|? ? 10M|? ?? ? |? ?162K??(1)| 00:32:36|

|? ?1 |??HASH GROUP BY? ?? ? |? ?? ?? ?? ? |? ?437K|? ? 10M|? ? 15M|? ?162K??(1)| 00:32:36|

|*??2 |? ?TABLE ACCESS FULL??| PROD_SEND? ?|? ?437K|? ? 10M|? ?? ? |? ?159K??(1)| 00:31:59|

------------------------------------------------------------------------------------------

Predicate Information (identified by operation id):

---------------------------------------------------

2 - filter("SCAN_TIME">=TO_DATE(' 2015-05-15 00:00:00', 'syyyy-mm-dd

hh24:mi:ss') AND "SCAN_TIME"

hh24:mi:ss'))

Statistics

----------------------------------------------------------

0??recursive calls

0??db block gets

586772??consistent gets

586762??physical reads

0??redo size

25943493??bytes sent via SQL*Net to client

624455??bytes received via SQL*Net from client

56723??SQL*Net roundtrips to/from client

0??sorts (memory)

0??sorts (disk)

850824??rows processed

總結

以上是生活随笔為你收集整理的Oracle distinct后加as,【大话IT】为何加distinct之后就不走索引了的全部內容，希望文章能夠幫你解決所遇到的問題。

如果覺得生活随笔網站內容還不錯，歡迎將生活随笔推薦給好友。