Oracle rownum影响运行计划
今天调优一条SQL语句，因为SQL比較复杂，用autotrace非常难一眼看出哪里出了问题，直接上10046。
SELECT AB.*
FROM (SELECT A.*, rownum RN
FROM (SELECT *
from (SELECT DISTINCT (D.DEVICE_ID), F.FUNCTION_LOCATION_ID
from GG_device D,
GG_CLASSIFY_CARD C,
GG_function_location F,
GG_fl_device L,
GG_tech_object_node n,
(SELECT N.TECH_OBJECT_ID
FROM GG_TECH_OBJECT_NODE N
WHERE N.NODE_TYPE = 2
START WITH N.TECH_OBJECT_ID = 15773325
CONNECT BY PRIOR
N.TECH_OBJECT_NODE_ID = N.PARENT_ID) TAB
where F.FUNCTION_LOCATION_ID = L.FUNCTION_LOCATION_ID
and L.Device_Id = d.device_id
and d.classify_id = c.classify_id
AND EXISTS
(SELECT 1
FROM GG_TECH_OBJECT_NODE N
WHERE N.TECH_OBJECT_ID = D.DEVICE_ID
AND N.NODE_TYPE = 2)
AND D.CURRENT_STATUS = 0
AND D.IS_SHARE_DEVICE = 1
AND TAB.TECH_OBJECT_ID = D.DEVICE_ID
and n.tech_object_id = f.function_location_id
AND F.SITE_ID = 1021
AND C.ALIAS_NAME IN ('A176')
union all
SELECT DISTINCT (D.DEVICE_ID), F.FUNCTION_LOCATION_ID
FROM GG_DEVICE D,
GG_CLASSIFY_CARD C,
GG_FUNCTION_LOCATION F,
GG_tech_object_node n,
(SELECT N.TECH_OBJECT_ID
FROM GG_TECH_OBJECT_NODE N
WHERE N.NODE_TYPE = 2
START WITH N.TECH_OBJECT_ID = 15773325
CONNECT BY PRIOR
N.TECH_OBJECT_NODE_ID = N.PARENT_ID) TAB
WHERE D.CLASSIFY_ID = C.CLASSIFY_ID
AND F.FUNCTION_LOCATION_ID(+) =
D.FUNCTION_LOCATION_ID
and n.tech_object_id = f.function_location_id
AND EXISTS
(SELECT 1
FROM GG_TECH_OBJECT_NODE N
WHERE N.TECH_OBJECT_ID = D.DEVICE_ID
AND N.NODE_TYPE = 2)
AND D.CURRENT_STATUS = 0
AND D.IS_SHARE_DEVICE = 0
AND TAB.TECH_OBJECT_ID = D.DEVICE_ID
AND F.SITE_ID = 1021
AND C.ALIAS_NAME IN ('A176'))) A
WHERE ROWNUM <= 25) AB
WHERE AB.RN > 0;
已用时间: 00: 00: 05.56
----------------------------------------------------------
Plan hash value: 1124467031
---------------------------------------------------------------------------------------------------------------------
| Id | Operation | Name | Rows | Bytes | Cost (%CPU)| Time |
---------------------------------------------------------------------------------------------------------------------
| 0 | SELECT STATEMENT | | 18 | 67248 | 28951 (1)| 00:05:48 |
|* 1 | VIEW | | 18 | 67248 | 28951 (1)| 00:05:48 |
|* 2 | COUNT STOPKEY | | | | | |
| 3 | VIEW | | 18 | 67014 | 28951 (1)| 00:05:48 |
|* 4 | SORT ORDER BY STOPKEY | | 18 | 67014 | 28951 (1)| 00:05:48 |
| 5 | VIEW | | 18 | 67014 | 28950 (1)| 00:05:48 |
| 6 | UNION-ALL | | | | | |
| 7 | HASH UNIQUE | | 6 | 2064 | 15146 (1)| 00:03:02 |
|* 8 | HASH JOIN | | 6 | 2064 | 15145 (1)| 00:03:02 |
| 9 | TABLE ACCESS BY INDEX ROWID | GG_TECH_OBJECT_NODE | 1 | 102 | 3 (0)| 00:00:01 |
| 10 | NESTED LOOPS | | 6 | 2022 | 5842 (1)| 00:01:11 |
| 11 | NESTED LOOPS SEMI | | 5 | 1175 | 5827 (1)| 00:01:10 |
| 12 | NESTED LOOPS | | 5 | 1125 | 5817 (1)| 00:01:10 |
|* 13 | HASH JOIN | | 172 | 32508 | 5645 (1)| 00:01:08 |
|* 14 | HASH JOIN | | 77 | 13629 | 5601 (1)| 00:01:08 |
|* 15 | TABLE ACCESS FULL | GG_CLASSIFY_CARD | 1 | 93 | 16 (0)| 00:00:01 |
|* 16 | TABLE ACCESS FULL | GG_DEVICE | 22527 | 1847K| 5584 (1)| 00:01:08 |
| 17 | TABLE ACCESS FULL | GG_FL_DEVICE | 74829 | 876K| 43 (3)| 00:00:01 |
|* 18 | TABLE ACCESS BY INDEX ROWID| GG_FUNCTION_LOCATION | 1 | 36 | 1 (0)| 00:00:01 |
|* 19 | INDEX UNIQUE SCAN | PK_GG_FUNCTION_LOCATION | 1 | | 0 (0)| 00:00:01 |
|* 20 | INDEX RANGE SCAN | IDX_TECH_NODE_ID | 482K| 4712K| 2 (0)| 00:00:01 |
|* 21 | INDEX RANGE SCAN | IDX_TECH_OBJECT_ID | 1 | | 2 (0)| 00:00:01 |
| 22 | VIEW | | 1762K| 11M| 9291 (1)| 00:01:52 |
|* 23 | FILTER | | | | | |
|* 24 | CONNECT BY WITH FILTERING | | | | | |
| 25 | TABLE ACCESS BY INDEX ROWID | GG_TECH_OBJECT_NODE | | | | |
|* 26 | INDEX RANGE SCAN | IDX_TECH_OBJECT_ID | 1 | 7 | 3 (0)| 00:00:01 |
|* 27 | HASH JOIN | | | | | |
| 28 | CONNECT BY PUMP | | | | | |
| 29 | TABLE ACCESS FULL | GG_TECH_OBJECT_NODE | 1762K| 40M| 9291 (1)| 00:01:52 |
| 30 | TABLE ACCESS FULL | GG_TECH_OBJECT_NODE | 1762K| 40M| 9291 (1)| 00:01:52 |
| 31 | HASH UNIQUE | | 12 | 4056 | 13804 (1)| 00:02:46 |
|* 32 | HASH JOIN | | 12 | 4056 | 13803 (1)| 00:02:46 |
| 33 | TABLE ACCESS BY INDEX ROWID | GG_TECH_OBJECT_NODE | 1 | 102 | 3 (0)| 00:00:01 |
| 34 | NESTED LOOPS | | 11 | 3641 | 4501 (1)| 00:00:55 |
| 35 | NESTED LOOPS SEMI | | 10 | 2290 | 4471 (1)| 00:00:54 |
|* 36 | HASH JOIN | | 10 | 2190 | 4451 (1)| 00:00:54 |
|* 37 | TABLE ACCESS FULL | GG_CLASSIFY_CARD | 1 | 93 | 16 (0)| 00:00:01 |
|* 38 | TABLE ACCESS BY INDEX ROWID| GG_DEVICE | 4 | 360 | 5 (0)| 00:00:01 |
| 39 | NESTED LOOPS | | 2823 | 347K| 4434 (1)| 00:00:54 |
|* 40 | TABLE ACCESS FULL | GG_FUNCTION_LOCATION | 685 | 24660 | 2214 (1)| 00:00:27 |
|* 41 | INDEX RANGE SCAN | IDX_FLOCID | 4 | | 2 (0)| 00:00:01 |
|* 42 | INDEX RANGE SCAN | IDX_TECH_NODE_ID | 482K| 4712K| 2 (0)| 00:00:01 |
|* 43 | INDEX RANGE SCAN | IDX_TECH_OBJECT_ID | 1 | | 2 (0)| 00:00:01 |
| 44 | VIEW | | 1762K| 11M| 9291 (1)| 00:01:52 |
|* 45 | FILTER | | | | | |
|* 46 | CONNECT BY WITH FILTERING | | | | | |
| 47 | TABLE ACCESS BY INDEX ROWID | GG_TECH_OBJECT_NODE | | | | |
|* 48 | INDEX RANGE SCAN | IDX_TECH_OBJECT_ID | 1 | 7 | 3 (0)| 00:00:01 |
|* 49 | HASH JOIN | | | | | |
| 50 | CONNECT BY PUMP | | | | | |
| 51 | TABLE ACCESS FULL | GG_TECH_OBJECT_NODE | 1762K| 40M| 9291 (1)| 00:01:52 |
| 52 | TABLE ACCESS FULL | GG_TECH_OBJECT_NODE | 1762K| 40M| 9291 (1)| 00:01:52 |
---------------------------------------------------------------------------------------------------------------------
Predicate Information (identified by operation id):
---------------------------------------------------
1 - filter("AB"."RN">0)
2 - filter(ROWNUM<=25)
4 - filter(ROWNUM<=25)
8 - access("TAB"."TECH_OBJECT_ID"="D"."DEVICE_ID")
13 - access("L"."DEVICE_ID"="D"."DEVICE_ID")
14 - access("D"."CLASSIFY_ID"="C"."CLASSIFY_ID")
15 - filter("C"."ALIAS_NAME"='A176')
16 - filter("D"."IS_SHARE_DEVICE"=1 AND "D"."CURRENT_STATUS"=0)
18 - filter("F"."SITE_ID"=1021)
19 - access("F"."FUNCTION_LOCATION_ID"="L"."FUNCTION_LOCATION_ID")
20 - access("N"."TECH_OBJECT_ID"="D"."DEVICE_ID" AND "N"."NODE_TYPE"=2)
21 - access("N"."TECH_OBJECT_ID"="F"."FUNCTION_LOCATION_ID")
23 - filter("N"."NODE_TYPE"=2)
24 - filter("N"."TECH_OBJECT_ID"=15773325)
26 - access("N"."TECH_OBJECT_ID"=15773325)
27 - access("N"."PARENT_ID"=NULL)
32 - access("TAB"."TECH_OBJECT_ID"="D"."DEVICE_ID")
36 - access("D"."CLASSIFY_ID"="C"."CLASSIFY_ID")
37 - filter("C"."ALIAS_NAME"='A176')
38 - filter("D"."IS_SHARE_DEVICE"=0 AND "D"."CURRENT_STATUS"=0)
40 - filter("F"."SITE_ID"=1021)
41 - access("F"."FUNCTION_LOCATION_ID"="D"."FUNCTION_LOCATION_ID")
42 - access("N"."TECH_OBJECT_ID"="D"."DEVICE_ID" AND "N"."NODE_TYPE"=2)
43 - access("N"."TECH_OBJECT_ID"="F"."FUNCTION_LOCATION_ID")
45 - filter("N"."NODE_TYPE"=2)
46 - filter("N"."TECH_OBJECT_ID"=15773325)
48 - access("N"."TECH_OBJECT_ID"=15773325)
49 - access("N"."PARENT_ID"=NULL)
统计信息
----------------------------------------------------------
9 recursive calls
0 db block gets
209163 consistent gets
0 physical reads
0 redo size
2890 bytes sent via SQL*Net to client
10811 bytes received via SQL*Net from client
2 SQL*Net roundtrips to/from client
63 sorts (memory)
0 sorts (disk)
1 rows processed
10046 trace的结果，能够一眼看出是递归出了问题，依照业务上来说一个节点以下没有多少数据啊，怎么会不走索引呢？
Rows Row Source Operation
------- ---------------------------------------------------
1 VIEW (cr=209038 pr=0 pw=0 time=5665797 us)
1 COUNT STOPKEY (cr=209038 pr=0 pw=0 time=5665774 us)
1 VIEW (cr=209038 pr=0 pw=0 time=5665768 us)
1 SORT ORDER BY STOPKEY (cr=209038 pr=0 pw=0 time=5665745 us)
1 VIEW (cr=209038 pr=0 pw=0 time=5665649 us)
1 UNION-ALL (cr=209038 pr=0 pw=0 time=5665631 us)
0 HASH UNIQUE (cr=25205 pr=0 pw=0 time=570538 us)
0 HASH JOIN (cr=25205 pr=0 pw=0 time=570360 us)
0 TABLE ACCESS BY INDEX ROWID GG_TECH_OBJECT_NODE (cr=25205 pr=0 pw=0 time=570095 us)
1 NESTED LOOPS (cr=25205 pr=0 pw=0 time=570075 us)
0 NESTED LOOPS SEMI (cr=25205 pr=0 pw=0 time=570069 us)
0 NESTED LOOPS (cr=25205 pr=0 pw=0 time=570066 us)
0 HASH JOIN (cr=25205 pr=0 pw=0 time=570062 us)
0 HASH JOIN (cr=25205 pr=0 pw=0 time=569617 us)
1 TABLE ACCESS FULL GG_CLASSIFY_CARD (cr=68 pr=0 pw=0 time=1467 us)
21206 TABLE ACCESS FULL GG_DEVICE (cr=25137 pr=0 pw=0 time=424214 us)
0 TABLE ACCESS FULL GG_FL_DEVICE (cr=0 pr=0 pw=0 time=0 us)
0 TABLE ACCESS BY INDEX ROWID GG_FUNCTION_LOCATION (cr=0 pr=0 pw=0 time=0 us)
0 INDEX UNIQUE SCAN PK_GG_FUNCTION_LOCATION (cr=0 pr=0 pw=0 time=0 us)(object id 508068)
0 INDEX RANGE SCAN IDX_TECH_NODE_ID (cr=0 pr=0 pw=0 time=0 us)(object id 541613)
0 INDEX RANGE SCAN IDX_TECH_OBJECT_ID (cr=0 pr=0 pw=0 time=0 us)(object id 508645)
0 VIEW (cr=0 pr=0 pw=0 time=0 us)
0 FILTER (cr=0 pr=0 pw=0 time=0 us)
0 CONNECT BY WITH FILTERING (cr=0 pr=0 pw=0 time=0 us)
0 TABLE ACCESS BY INDEX ROWID GG_TECH_OBJECT_NODE (cr=0 pr=0 pw=0 time=0 us)
0 INDEX RANGE SCAN IDX_TECH_OBJECT_ID (cr=0 pr=0 pw=0 time=0 us)(object id 508645)
0 HASH JOIN (cr=0 pr=0 pw=0 time=0 us)
0 CONNECT BY PUMP (cr=0 pr=0 pw=0 time=0 us)
0 TABLE ACCESS FULL GG_TECH_OBJECT_NODE (cr=0 pr=0 pw=0 time=0 us)
0 TABLE ACCESS FULL GG_TECH_OBJECT_NODE (cr=0 pr=0 pw=0 time=0 us)
1 HASH UNIQUE (cr=183833 pr=0 pw=0 time=5095035 us)
1 HASH JOIN (cr=183748 pr=0 pw=0 time=5090111 us)
170 TABLE ACCESS BY INDEX ROWID GG_TECH_OBJECT_NODE (cr=15772 pr=0 pw=0 time=7653 us)
341 NESTED LOOPS (cr=15610 pr=0 pw=0 time=189743 us)
170 NESTED LOOPS SEMI (cr=15268 pr=0 pw=0 time=5170 us)
170 HASH JOIN (cr=14926 pr=0 pw=0 time=3232 us)
1 TABLE ACCESS FULL GG_CLASSIFY_CARD (cr=68 pr=0 pw=0 time=830 us)
2867 TABLE ACCESS BY INDEX ROWID GG_DEVICE (cr=14858 pr=0 pw=0 time=28976 us)
4728 NESTED LOOPS (cr=13282 pr=0 pw=0 time=94565 us)
1667 TABLE ACCESS FULL GG_FUNCTION_LOCATION (cr=9937 pr=0 pw=0 time=13539 us)
3060 INDEX RANGE SCAN IDX_FLOCID (cr=3345 pr=0 pw=0 time=12458 us)(object id 507929)
170 INDEX RANGE SCAN IDX_TECH_NODE_ID (cr=342 pr=0 pw=0 time=3171 us)(object id 541613)
170 INDEX RANGE SCAN IDX_TECH_OBJECT_ID (cr=342 pr=0 pw=0 time=1854 us)(object id 508645)
31 VIEW (cr=167976 pr=0 pw=0 time=4864861 us)
31 FILTER (cr=167976 pr=0 pw=0 time=4864794 us)
56 CONNECT BY WITH FILTERING (cr=167976 pr=0 pw=0 time=4865653 us)
1 TABLE ACCESS BY INDEX ROWID GG_TECH_OBJECT_NODE (cr=4 pr=0 pw=0 time=58 us)
1 INDEX RANGE SCAN IDX_TECH_OBJECT_ID (cr=3 pr=0 pw=0 time=25 us)(object id 508645)
55 HASH JOIN (cr=167972 pr=0 pw=0 time=1264029 us)
56 CONNECT BY PUMP (cr=0 pr=0 pw=0 time=56 us)
7048956 TABLE ACCESS FULL GG_TECH_OBJECT_NODE (cr=167972 pr=0 pw=0 time=488 us)
0 TABLE ACCESS FULL GG_TECH_OBJECT_NODE (cr=0 pr=0 pw=0 time=0 us)
在无意中測试发现，去掉分页的语句之后，就很快了。揣測是COUNT STOPKEY造成的，假设数据量大的做分页，肯定是能够提升性能的，但此条SQL语句仅仅是返回一条数据。我的结论是rownum能够改变运行计划。
SQL> SELECT *
from (SELECT DISTINCT (D.DEVICE_ID), F.FUNCTION_LOCATION_ID
from GG_device D,
GG_CLASSIFY_CARD C,
GG_function_location F,
GG_fl_device L,
GG_tech_object_node n,
(SELECT N.TECH_OBJECT_ID
FROM GG_TECH_OBJECT_NODE N
WHERE N.NODE_TYPE = 2
START WITH N.TECH_OBJECT_ID = 15773325
CONNECT BY PRIOR N.TECH_OBJECT_NODE_ID = N.PARENT_ID) TAB
where F.FUNCTION_LOCATION_ID = L.FUNCTION_LOCATION_ID
and L.Device_Id = d.device_id
and d.classify_id = c.classify_id
AND EXISTS (SELECT 1
FROM GG_TECH_OBJECT_NODE N
WHERE N.TECH_OBJECT_ID = D.DEVICE_ID
AND N.NODE_TYPE = 2)
AND D.CURRENT_STATUS = 0
AND D.IS_SHARE_DEVICE = 1
AND TAB.TECH_OBJECT_ID = D.DEVICE_ID
and n.tech_object_id = f.function_location_id
AND F.SITE_ID = 1021
AND C.ALIAS_NAME IN ('A176')
union all
SELECT DISTINCT (D.DEVICE_ID), F.FUNCTION_LOCATION_ID
FROM GG_DEVICE D,
GG_CLASSIFY_CARD C,
GG_FUNCTION_LOCATION F,
GG_tech_object_node n,
(SELECT N.TECH_OBJECT_ID
FROM GG_TECH_OBJECT_NODE N
WHERE N.NODE_TYPE = 2
START WITH N.TECH_OBJECT_ID = 15773325
CONNECT BY PRIOR N.TECH_OBJECT_NODE_ID = N.PARENT_ID) TAB
WHERE D.CLASSIFY_ID = C.CLASSIFY_ID
AND F.FUNCTION_LOCATION_ID(+) = D.FUNCTION_LOCATION_ID
and n.tech_object_id = f.function_location_id
AND EXISTS (SELECT 1
FROM GG_TECH_OBJECT_NODE N
WHERE N.TECH_OBJECT_ID = D.DEVICE_ID
AND N.NODE_TYPE = 2)
AND D.CURRENT_STATUS = 0
AND D.IS_SHARE_DEVICE = 0
AND TAB.TECH_OBJECT_ID = D.DEVICE_ID
AND F.SITE_ID = 1021
AND C.ALIAS_NAME IN ('A176'));
已用时间: 00: 00: 00.06
运行计划
----------------------------------------------------------
Plan hash value: 1345020195
------------------------------------------------------------------------------------------------------------------------
| Id | Operation | Name | Rows | Bytes | Cost (%CPU)| Time |
------------------------------------------------------------------------------------------------------------------------
| 0 | SELECT STATEMENT | | 2 | 7446 | 41 (5)| 00:00:01 |
| 1 | VIEW | | 2 | 7446 | 41 (5)| 00:00:01 |
| 2 | UNION-ALL | | | | | |
| 3 | HASH UNIQUE | | 1 | 345 | 21 (5)| 00:00:01 |
| 4 | TABLE ACCESS BY INDEX ROWID | GG_TECH_OBJECT_NODE | 1 | 103 | 3 (0)| 00:00:01 |
| 5 | NESTED LOOPS | | 1 | 345 | 20 (0)| 00:00:01 |
| 6 | NESTED LOOPS | | 1 | 242 | 17 (0)| 00:00:01 |
| 7 | NESTED LOOPS | | 1 | 206 | 16 (0)| 00:00:01 |
| 8 | NESTED LOOPS SEMI | | 1 | 194 | 15 (0)| 00:00:01 |
| 9 | NESTED LOOPS | | 1 | 184 | 13 (0)| 00:00:01 |
| 10 | NESTED LOOPS | | 3 | 273 | 10 (0)| 00:00:01 |
| 11 | VIEW | | 3 | 21 | 6 (0)| 00:00:01 |
|* 12 | FILTER | | | | | |
|* 13 | CONNECT BY WITH FILTERING | | | | | |
| 14 | TABLE ACCESS BY INDEX ROWID | GG_TECH_OBJECT_NODE | | | | |
|* 15 | INDEX RANGE SCAN | IDX_TECH_OBJECT_ID1 | 1 | 7 | 3 (0)| 00:00:01 |
| 16 | NESTED LOOPS | | | | | |
| 17 | BUFFER SORT | | | | | |
| 18 | CONNECT BY PUMP | | | | | |
| 19 | TABLE ACCESS BY INDEX ROWID| GG_TECH_OBJECT_NODE | 3 | 72 | 6 (0)| 00:00:01 |
|* 20 | INDEX RANGE SCAN | IDX_TECH_OBJECT_PARENT_ID1 | 3 | | 3 (0)| 00:00:01 |
|* 21 | TABLE ACCESS FULL | GG_TECH_OBJECT_NODE | 3 | 72 | 6 (0)| 00:00:01 |
|* 22 | TABLE ACCESS BY INDEX ROWID | GG_DEVICE | 1 | 84 | 2 (0)| 00:00:01 |
|* 23 | INDEX UNIQUE SCAN | PK_GG_DEVICE | 1 | | 1 (0)| 00:00:01 |
|* 24 | TABLE ACCESS BY INDEX ROWID | GG_CLASSIFY_CARD | 1 | 93 | 1 (0)| 00:00:01 |
|* 25 | INDEX UNIQUE SCAN | SYS_C00468549 | 1 | | 0 (0)| 00:00:01 |
|* 26 | INDEX RANGE SCAN | IDX_TECH_NODE_ID1 | 587K| 5736K| 2 (0)| 00:00:01 |
|* 27 | INDEX RANGE SCAN | PK_GG_FL_DEVICE | 2 | 24 | 1 (0)| 00:00:01 |
|* 28 | TABLE ACCESS BY INDEX ROWID | GG_FUNCTION_LOCATION | 1 | 36 | 1 (0)| 00:00:01 |
|* 29 | INDEX UNIQUE SCAN | PK_GG_FUNCTION_LOCATION | 1 | | 0 (0)| 00:00:01 |
|* 30 | INDEX RANGE SCAN | IDX_TECH_OBJECT_ID1 | 1 | | 2 (0)| 00:00:01 |
| 31 | HASH UNIQUE | | 1 | 339 | 20 (5)| 00:00:01 |
| 32 | TABLE ACCESS BY INDEX ROWID | GG_TECH_OBJECT_NODE | 1 | 103 | 3 (0)| 00:00:01 |
| 33 | NESTED LOOPS | | 1 | 339 | 19 (0)| 00:00:01 |
| 34 | NESTED LOOPS SEMI | | 1 | 236 | 16 (0)| 00:00:01 |
| 35 | NESTED LOOPS | | 1 | 226 | 14 (0)| 00:00:01 |
| 36 | NESTED LOOPS | | 1 | 133 | 13 (0)| 00:00:01 |
| 37 | NESTED LOOPS | | 3 | 291 | 10 (0)| 00:00:01 |
| 38 | VIEW | | 3 | 21 | 6 (0)| 00:00:01 |
|* 39 | FILTER | | | | | |
|* 40 | CONNECT BY WITH FILTERING | | | | | |
| 41 | TABLE ACCESS BY INDEX ROWID | GG_TECH_OBJECT_NODE | | | | |
|* 42 | INDEX RANGE SCAN | IDX_TECH_OBJECT_ID1 | 1 | 7 | 3 (0)| 00:00:01 |
| 43 | NESTED LOOPS | | | | | |
| 44 | BUFFER SORT | | | | | |
| 45 | CONNECT BY PUMP | | | | | |
| 46 | TABLE ACCESS BY INDEX ROWID | GG_TECH_OBJECT_NODE | 3 | 72 | 6 (0)| 00:00:01 |
|* 47 | INDEX RANGE SCAN | IDX_TECH_OBJECT_PARENT_ID1 | 3 | | 3 (0)| 00:00:01 |
|* 48 | TABLE ACCESS FULL | GG_TECH_OBJECT_NODE | 3 | 72 | 6 (0)| 00:00:01 |
|* 49 | TABLE ACCESS BY INDEX ROWID | GG_DEVICE | 1 | 90 | 2 (0)| 00:00:01 |
|* 50 | INDEX UNIQUE SCAN | PK_GG_DEVICE | 1 | | 1 (0)| 00:00:01 |
|* 51 | TABLE ACCESS BY INDEX ROWID | GG_FUNCTION_LOCATION | 1 | 36 | 1 (0)| 00:00:01 |
|* 52 | INDEX UNIQUE SCAN | PK_GG_FUNCTION_LOCATION | 1 | | 0 (0)| 00:00:01 |
|* 53 | TABLE ACCESS BY INDEX ROWID | GG_CLASSIFY_CARD | 1 | 93 | 1 (0)| 00:00:01 |
|* 54 | INDEX UNIQUE SCAN | SYS_C00468549 | 1 | | 0 (0)| 00:00:01 |
|* 55 | INDEX RANGE SCAN | IDX_TECH_NODE_ID1 | 587K| 5736K| 2 (0)| 00:00:01 |
|* 56 | INDEX RANGE SCAN | IDX_TECH_OBJECT_ID1 | 1 | | 2 (0)| 00:00:01 |
------------------------------------------------------------------------------------------------------------------------
Predicate Information (identified by operation id):
---------------------------------------------------
12 - filter("N"."NODE_TYPE"=2)
13 - filter("N"."TECH_OBJECT_ID"=15773325)
15 - access("N"."TECH_OBJECT_ID"=15773325)
20 - access("N"."PARENT_ID"=NULL)
21 - access("N"."PARENT_ID"=NULL)
22 - filter("D"."IS_SHARE_DEVICE"=1 AND "D"."CURRENT_STATUS"=0)
23 - access("TAB"."TECH_OBJECT_ID"="D"."DEVICE_ID")
24 - filter("C"."ALIAS_NAME"='A176')
25 - access("D"."CLASSIFY_ID"="C"."CLASSIFY_ID")
26 - access("N"."TECH_OBJECT_ID"="D"."DEVICE_ID" AND "N"."NODE_TYPE"=2)
27 - access("L"."DEVICE_ID"="D"."DEVICE_ID")
28 - filter("F"."SITE_ID"=1021)
29 - access("F"."FUNCTION_LOCATION_ID"="L"."FUNCTION_LOCATION_ID")
30 - access("N"."TECH_OBJECT_ID"="F"."FUNCTION_LOCATION_ID")
39 - filter("N"."NODE_TYPE"=2)
40 - filter("N"."TECH_OBJECT_ID"=15773325)
42 - access("N"."TECH_OBJECT_ID"=15773325)
47 - access("N"."PARENT_ID"=NULL)
48 - access("N"."PARENT_ID"=NULL)
49 - filter("D"."IS_SHARE_DEVICE"=0 AND "D"."CURRENT_STATUS"=0)
50 - access("TAB"."TECH_OBJECT_ID"="D"."DEVICE_ID")
51 - filter("F"."SITE_ID"=1021)
52 - access("F"."FUNCTION_LOCATION_ID"="D"."FUNCTION_LOCATION_ID")
53 - filter("C"."ALIAS_NAME"='A176')
54 - access("D"."CLASSIFY_ID"="C"."CLASSIFY_ID")
55 - access("N"."TECH_OBJECT_ID"="D"."DEVICE_ID" AND "N"."NODE_TYPE"=2)
56 - access("N"."TECH_OBJECT_ID"="F"."FUNCTION_LOCATION_ID")
统计信息
----------------------------------------------------------
7 recursive calls
0 db block gets
738 consistent gets
0 physical reads
0 redo size
2606 bytes sent via SQL*Net to client
10273 bytes received via SQL*Net from client
2 SQL*Net roundtrips to/from client
18 sorts (memory)
0 sorts (disk)
1 rows processed
用还有一种方式实现分页。
SELECT AB.*
FROM (SELECT tt.*,row_number()over(ORDER BY SORT_NO, name) nr
from (SELECT DISTINCT (D.DEVICE_ID),
D.NAME,
N.Full_Path SORT_NO,
F.FUNCTION_LOCATION_ID
from GG_device D,
GG_CLASSIFY_CARD C,
GG_function_location F,
GG_fl_device L,
GG_tech_object_node n,
(SELECT N.TECH_OBJECT_ID
FROM GG_TECH_OBJECT_NODE N
WHERE N.NODE_TYPE = 2
START WITH N.TECH_OBJECT_ID = 15773325
CONNECT BY PRIOR
N.TECH_OBJECT_NODE_ID = N.PARENT_ID) TAB
where F.FUNCTION_LOCATION_ID =
L.FUNCTION_LOCATION_ID
and L.Device_Id = d.device_id
and d.classify_id = c.classify_id
AND EXISTS
(SELECT 1
FROM GG_TECH_OBJECT_NODE N
WHERE N.TECH_OBJECT_ID = D.DEVICE_ID
AND N.NODE_TYPE = 2)
AND D.CURRENT_STATUS = 0
AND D.IS_SHARE_DEVICE = 1
AND TAB.TECH_OBJECT_ID = D.DEVICE_ID
and n.tech_object_id = f.function_location_id
AND F.SITE_ID = 1021
AND C.ALIAS_NAME IN ('A176')
union all
SELECT DISTINCT (D.DEVICE_ID),
D.NAME,
N.Full_Path SORT_NO,
F.FUNCTION_LOCATION_ID
FROM GG_DEVICE D,
GG_CLASSIFY_CARD C,
GG_FUNCTION_LOCATION F,
GG_tech_object_node n,
(SELECT N.TECH_OBJECT_ID
FROM GG_TECH_OBJECT_NODE N
WHERE N.NODE_TYPE = 2
START WITH N.TECH_OBJECT_ID = 15773325
CONNECT BY PRIOR
N.TECH_OBJECT_NODE_ID = N.PARENT_ID) TAB
WHERE D.CLASSIFY_ID = C.CLASSIFY_ID
AND F.FUNCTION_LOCATION_ID(+) =
D.FUNCTION_LOCATION_ID
and n.tech_object_id = f.function_location_id
AND EXISTS
(SELECT 1
FROM GG_TECH_OBJECT_NODE N
WHERE N.TECH_OBJECT_ID = D.DEVICE_ID
AND N.NODE_TYPE = 2)
AND D.CURRENT_STATUS = 0
AND D.IS_SHARE_DEVICE = 0
AND TAB.TECH_OBJECT_ID = D.DEVICE_ID
AND F.SITE_ID = 1021
AND C.ALIAS_NAME IN ('A176')) tt
) AB where ab.nr between 1 and 25;
已用时间: 00: 00: 00.06
运行计划
----------------------------------------------------------
Plan hash value: 3880620066
--------------------------------------------------------------------------------------------------------------------------
| Id | Operation | Name | Rows | Bytes | Cost (%CPU)| Time |
--------------------------------------------------------------------------------------------------------------------------
| 0 | SELECT STATEMENT | | 2 | 7472 | 42 (8)| 00:00:01 |
|* 1 | VIEW | | 2 | 7472 | 42 (8)| 00:00:01 |
|* 2 | WINDOW SORT PUSHED RANK | | 2 | 7446 | 42 (8)| 00:00:01 |
| 3 | VIEW | | 2 | 7446 | 41 (5)| 00:00:01 |
| 4 | UNION-ALL | | | | | |
| 5 | HASH UNIQUE | | 1 | 345 | 21 (5)| 00:00:01 |
| 6 | TABLE ACCESS BY INDEX ROWID | GG_TECH_OBJECT_NODE | 1 | 103 | 3 (0)| 00:00:01 |
| 7 | NESTED LOOPS | | 1 | 345 | 20 (0)| 00:00:01 |
| 8 | NESTED LOOPS | | 1 | 242 | 17 (0)| 00:00:01 |
| 9 | NESTED LOOPS | | 1 | 206 | 16 (0)| 00:00:01 |
| 10 | NESTED LOOPS SEMI | | 1 | 194 | 15 (0)| 00:00:01 |
| 11 | NESTED LOOPS | | 1 | 254 | 13 (0)| 00:00:01 |
| 12 | NESTED LOOPS | | 3 | 273 | 10 (0)| 00:00:01 |
| 13 | VIEW | | 3 | 21 | 6 (0)| 00:00:01 |
|* 14 | FILTER | | | | | |
|* 15 | CONNECT BY WITH FILTERING | | | | | |
| 16 | TABLE ACCESS BY INDEX ROWID | GG_TECH_OBJECT_NODE | | | | |
|* 17 | INDEX RANGE SCAN | IDX_TECH_OBJECT_ID1 | 1 | 7 | 3 (0)| 00:00:01 |
| 25 | NESTED LOOPS | | | | | |
| 19 | BUFFER SORT | | | | | |
| 20 | CONNECT BY PUMP | | | | | |
| 21 | TABLE ACCESS BY INDEX ROWID| GG_TECH_OBJECT_NODE | 3 | 72 | 6 (0)| 00:00:01 |
|* 22 | INDEX RANGE SCAN | IDX_TECH_OBJECT_PARENT_ID1 | 3 | | 3 (0)| 00:00:01 |
|* 23 | TABLE ACCESS FULL | GG_TECH_OBJECT_NODE | 3 | 72 | 6 (0)| 00:00:01 |
|* 24 | TABLE ACCESS BY INDEX ROWID | GG_DEVICE | 1 | 84 | 2 (0)| 00:00:01 |
|* 25 | INDEX UNIQUE SCAN | PK_GG_DEVICE | 1 | | 1 (0)| 00:00:01 |
|* 26 | TABLE ACCESS BY INDEX ROWID | GG_CLASSIFY_CARD | 1 | 93 | 1 (0)| 00:00:01 |
|* 27 | INDEX UNIQUE SCAN | SYS_C00468549 | 1 | | 0 (0)| 00:00:01 |
|* 28 | INDEX RANGE SCAN | IDX_TECH_NODE_ID1 | 587K| 5736K| 2 (0)| 00:00:01 |
|* 29 | INDEX RANGE SCAN | PK_GG_FL_DEVICE | 2 | 24 | 1 (0)| 00:00:01 |
|* 30 | TABLE ACCESS BY INDEX ROWID | GG_FUNCTION_LOCATION | 1 | 36 | 1 (0)| 00:00:01 |
|* 31 | INDEX UNIQUE SCAN | PK_GG_FUNCTION_LOCATION | 1 | | 0 (0)| 00:00:01 |
|* 32 | INDEX RANGE SCAN | IDX_TECH_OBJECT_ID1 | 1 | | 2 (0)| 00:00:01 |
| 33 | HASH UNIQUE | | 1 | 339 | 20 (5)| 00:00:01 |
| 34 | TABLE ACCESS BY INDEX ROWID | GG_TECH_OBJECT_NODE | 1 | 103 | 3 (0)| 00:00:01 |
| 35 | NESTED LOOPS | | 1 | 339 | 19 (0)| 00:00:01 |
| 36 | NESTED LOOPS SEMI | | 1 | 236 | 16 (0)| 00:00:01 |
| 37 | NESTED LOOPS | | 1 | 226 | 14 (0)| 00:00:01 |
| 38 | NESTED LOOPS | | 1 | 133 | 13 (0)| 00:00:01 |
| 39 | NESTED LOOPS | | 3 | 291 | 10 (0)| 00:00:01 |
| 40 | VIEW | | 3 | 21 | 6 (0)| 00:00:01 |
|* 41 | FILTER | | | | | |
|* 42 | CONNECT BY WITH FILTERING | | | | | |
| 43 | TABLE ACCESS BY INDEX ROWID | GG_TECH_OBJECT_NODE | | | | |
|* 44 | INDEX RANGE SCAN | IDX_TECH_OBJECT_ID1 | 1 | 7 | 3 (0)| 00:00:01 |
| 45 | NESTED LOOPS | | | | | |
| 46 | BUFFER SORT | | | | | |
| 47 | CONNECT BY PUMP | | | | | |
| 48 | TABLE ACCESS BY INDEX ROWID | GG_TECH_OBJECT_NODE | 3 | 72 | 6 (0)| 00:00:01 |
|* 49 | INDEX RANGE SCAN | IDX_TECH_OBJECT_PARENT_ID1 | 3 | | 3 (0)| 00:00:01 |
|* 50 | TABLE ACCESS FULL | GG_TECH_OBJECT_NODE | 3 | 72 | 6 (0)| 00:00:01 |
|* 51 | TABLE ACCESS BY INDEX ROWID | GG_DEVICE | 1 | 90 | 2 (0)| 00:00:01 |
|* 52 | INDEX UNIQUE SCAN | PK_GG_DEVICE | 1 | | 1 (0)| 00:00:01 |
|* 53 | TABLE ACCESS BY INDEX ROWID | GG_FUNCTION_LOCATION | 1 | 36 | 1 (0)| 00:00:01 |
|* 54 | INDEX UNIQUE SCAN | PK_GG_FUNCTION_LOCATION | 1 | | 0 (0)| 00:00:01 |
|* 55 | TABLE ACCESS BY INDEX ROWID | GG_CLASSIFY_CARD | 1 | 93 | 1 (0)| 00:00:01 |
|* 56 | INDEX UNIQUE SCAN | SYS_C00468549 | 1 | | 0 (0)| 00:00:01 |
|* 57 | INDEX RANGE SCAN | IDX_TECH_NODE_ID1 | 587K| 5736K| 2 (0)| 00:00:01 |
|* 58 | INDEX RANGE SCAN | IDX_TECH_OBJECT_ID1 | 1 | | 2 (0)| 00:00:01 |
--------------------------------------------------------------------------------------------------------------------------
Predicate Information (identified by operation id):
---------------------------------------------------
1 - filter("AB"."NR">=1 AND "AB"."NR"<=25)
2 - filter(ROW_NUMBER() OVER ( ORDER BY "SORT_NO","NAME")<=25)
14 - filter("N"."NODE_TYPE"=2 AND 1<=25)
15 - filter("N"."TECH_OBJECT_ID"=15773325)
17 - access("N"."TECH_OBJECT_ID"=15773325)
22 - access("N"."PARENT_ID"=NULL)
23 - access("N"."PARENT_ID"=NULL)
24 - filter("D"."IS_SHARE_DEVICE"=1 AND "D"."CURRENT_STATUS"=0)
25 - access("TAB"."TECH_OBJECT_ID"="D"."DEVICE_ID")
26 - filter("C"."ALIAS_NAME"='A176')
27 - access("D"."CLASSIFY_ID"="C"."CLASSIFY_ID")
28 - access("N"."TECH_OBJECT_ID"="D"."DEVICE_ID" AND "N"."NODE_TYPE"=2)
29 - access("L"."DEVICE_ID"="D"."DEVICE_ID")
30 - filter("F"."SITE_ID"=1021)
31 - access("F"."FUNCTION_LOCATION_ID"="L"."FUNCTION_LOCATION_ID")
32 - access("N"."TECH_OBJECT_ID"="F"."FUNCTION_LOCATION_ID")
41 - filter("N"."NODE_TYPE"=2 AND 1<=25)
42 - filter("N"."TECH_OBJECT_ID"=15773325)
44 - access("N"."TECH_OBJECT_ID"=15773325)
49 - access("N"."PARENT_ID"=NULL)
50 - access("N"."PARENT_ID"=NULL)
51 - filter("D"."IS_SHARE_DEVICE"=0 AND "D"."CURRENT_STATUS"=0)
52 - access("TAB"."TECH_OBJECT_ID"="D"."DEVICE_ID")
53 - filter("F"."SITE_ID"=1021)
54 - access("F"."FUNCTION_LOCATION_ID"="D"."FUNCTION_LOCATION_ID")
55 - filter("C"."ALIAS_NAME"='A176')
56 - access("D"."CLASSIFY_ID"="C"."CLASSIFY_ID")
57 - access("N"."TECH_OBJECT_ID"="D"."DEVICE_ID" AND "N"."NODE_TYPE"=2)
58 - access("N"."TECH_OBJECT_ID"="F"."FUNCTION_LOCATION_ID")
统计信息
----------------------------------------------------------
8 recursive calls
0 db block gets
863 consistent gets
0 physical reads
0 redo size
2890 bytes sent via SQL*Net to client
10807 bytes received via SQL*Net from client
2 SQL*Net roundtrips to/from client
76 sorts (memory)
0 sorts (disk)
1 rows processed
Oracle rownum影响运行计划的更多相关文章
- 使用hint优化Oracle的运行计划 以及 SQL Tune Advisor的使用
背景: 某表忽然出现查询很缓慢的情况.cost 100+ 秒以上:严重影响生产. 原SQL: explain plan for select * from ( select ID id,RET_NO ...
- [Oracle] 获取运行计划的各方法总结
总的结论: 一.获取运行计划的6种方法(具体步骤已经在每一个样例的开头凝视部分说明了): 1. explain plan for获取: 2. set autotrace on . 3. stati ...
- Oracle 11g 递归+ exists运行计划的改变
有一个递归查询在10g上执行非常快,但在11g上执行不出来. SQL> select * from v$version; BANNER ----------------------------- ...
- oracle分区表运行计划
分区表有非常多优点,以大化小,一小化了,加上并行的使用,在loap中能往往能提高几十倍甚至几百倍的效果. 当然表设计得不好也会适得其反.效果比普通表跟糟糕. 为了更好的使用分区表,这里看一下分区表的运 ...
- Oracle中的执行计划
使用autotrace sqlplus系统参数:SQL> set autotrace trace onSQL> select * from dual;DUM---XExecution Pl ...
- MySQL运行计划初探
-Mysql运行计划总结– 1 运行计划概述 先看看一个运行计划 mysql> explain SELECT * FROM EMP , DAO_OBJECTS t1 , DAO_OBJECTS ...
- 闪回drop恢复表后sql运行计划异常
-----正常运行计划 set autotrace traceonly set linesize 1000 select /*+index(t idx_object_id)*/ * from t wh ...
- Spark入门实战系列--6.SparkSQL（中）--深入了解SparkSQL运行计划及调优
[注]该系列文章以及使用到安装包/测试数据 可以在<倾情大奉送--Spark入门实战系列>获取 1.1 运行环境说明 1.1.1 硬软件环境 线程,主频2.2G,10G内存 l 虚拟软 ...
- Oracle中获取执行计划的几种方法分析
以下是对Oracle中获取执行计划的几种方法进行了详细的分析介绍,需要的朋友可以参考下 1. 预估执行计划 - Explain PlanExplain plan以SQL语句作为输入,得到这条S ...
随机推荐
- 编写python代码获取4k高清壁纸
Huskiesir最近在研究python爬虫大约俩周了吧,由于比较懒,也没把具体研究的过程与经验写下来,实在是一大憾事.这次直接上干货,代码送给大家: import re import request ...
- DQL查询语句使用(select)
9)DQL查询语句使用 SELECT语句在PL/SQL中使用,必须 采用下面用法: select id INTO 变量 from t001 where id=5; 将记录字段 ...
- Python中常见的文件对象内建函数
文件对象内建方法列表 文件对象的方法 操作 file.close() 关闭文件 file.fileno() 返回文件的描写叙述符(file descriptor.FD,整数值) file.flush( ...
- How to get the MouseEvent coordinates for an element that has CSS3 Transform?
I want to detect where a MouseEvent has occurred, in coordinates relative to the clicked element. Wh ...
- node 内存消息队列
var net = require('net') var clients = [] ,msgs = {} function unWrapMsg(data){ data = data.toString( ...
- 项目复习期总结3：CSS引入方式，凝视，命名规范，背景，行高，文本属性
文件夹: 1.CSS凝视的书写 怎么写?优点? 2.CSS引入方式 各种的优缺点 3.选择器的写法与选择器的优先级 4.CSS命名规范 5.背景,行高 6.文本(text与font开头)等全部属性 ...
- Hive编程指南_学习笔记01
第四章: HQl的数据定义 1:创建数据库 create database financials; create database if not exists financials; 2: ...
- C++ 虚函数的缺省參数问题
前些日子,有个同学问我一个关于虚函数的缺省參数问题.他是从某个论坛上看到的.可是自己没想通.便来找我. 如今分享一下这个问题.先看一小段代码: #include <iostream> us ...
- JavaScript事件驱动机制&定时器机制
在浏览器中,事件作为一个极为重要的机制,给予JavaScript响应用户操作与DOM变化的能力.在NodeJS中.异步事件驱动模型则是提高并发能力的基础. 一.程序怎样响应事件 程序响应外部的事件有两 ...
- [JZOJ 5906] [NOIP2018模拟10.15] 传送门 解题报告（树形DP）
题目链接: https://jzoj.net/senior/#contest/show/2528/2 题目: 8102年,Normalgod在GLaDOS的帮助下,研制出了传送枪.但GLaDOS想把传 ...