视图合并、hash join连接列数据分布不均匀引发的惨案

表大小

1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
SQL> select count(*) from agent.TB_AGENT_INFO;
COUNT(*)
----------
1751
SQL> select count(*) from TB_CHANNEL_INFO ;
COUNT(*)
----------
1807
SQL> select count(*) from TB_USER_CHANNEL;
COUNT(*)
----------
7269
SQL> select count(*) from OSS_USER_STATION;
COUNT(*)
----------
2149
SQL> select count(*) from tb_user_zgy ;
COUNT(*)
----------
43
SQL> select count(*) from act.tb_user_agent_relat;
COUNT(*)
----------
29612
SQL> select count(*) from agent.base_data_user_info ;
COUNT(*)
----------
30005
SQL> select count(*) from agent.base_data_invest_info;
COUNT(*)
----------
3530163

慢的sql

1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
61
62
63
64
65
66
67
68
69
70
71
72
73
74
75
76
77
78
79
80
81
82
83
84
85
86
87
88
89
90
91
92
93
94
95
96
97
98
99
100
101
102
103
104
105
106
107
108
select a.city,
a.agent_id,
a.username,
a.real_name,
phone,
zgy_name,
login_count,
user_count,
count(distinct b.invest_id) user_invested,
sum(b.order_amount / 100) invest_amount
from (select a.city,
a.agent_id,
a.username,
a.real_name, -- 业主姓名
a.phone, -- 业主手机号
d.real_name zgy_name, -- 所属专管员
count(distinct case
when c.str_day <= '20160821' then
c.login_name
end) login_count,
count(distinct case
when c.str_day <= '20160821' then
decode(c.status, 1, c.invest_id, null)
end) user_count
from (select agent_id, city, username, real_name, phone
from agent.TB_AGENT_INFO
where agent_id in
(SELECT agent_id
FROM (SELECT distinct *
FROM TB_CHANNEL_INFO t
START WITH t.CHANNEL_ID in
(select CHANNEL_ID
from TB_USER_CHANNEL
where USER_ID = 596)
CONNECT BY PRIOR
t.CHANNEL_ID = t.PARENT_CHANNEL_ID)
WHERE agent_id IS NOT NULL)) a
left join oss_user_station e
on a.agent_id = e.agent_id
and e.user_type = 0
left join tb_user_zgy d
on e.username = d.username
left join act.tb_user_agent_relat c
on a.agent_id = c.agent_id
group by a.city,
a.username,
a.real_name,
a.phone,
d.real_name,
a.agent_id) a
left join (select invest_id, order_amount, agent_id, str_day
from agent.base_data_invest_info
where str_day >= '20150801' and str_day<='20160821') b
on a.agent_id = b.agent_id
group by a.city,
a.agent_id,
a.username,
a.real_name,
a.phone,
a.zgy_name,
a.login_count,
a.user_count
这个查询可以看成两部分,第一部分一堆小表关联的a和唯一的一个大表再做关联
man
----------------------------------------------------------------------------------------------------------------------------------
| Id | Operation | Name | Rows | Bytes |TempSpc| Cost (%CPU)| Time |
----------------------------------------------------------------------------------------------------------------------------------
| 0 | SELECT STATEMENT | | 55M| 6616M| | 3934K (1)| 13:06:52 |
| 1 | HASH GROUP BY | | 55M| 6616M| | 3934K (1)| 13:06:52 |
| 2 | VIEW | VW_DAG_1 | 55M| 6616M| | 3934K (1)| 13:06:52 |
| 3 | HASH GROUP BY | | 55M| 6301M| 7681M| 3934K (1)| 13:06:52 |
| 4 | VIEW | VM_NWVW_0 | 55M| 6301M| | 2456K (1)| 08:11:15 |
| 5 | SORT GROUP BY | | 55M| 10G| 11G| 2456K (1)| 08:11:15 |
|* 6 | HASH JOIN RIGHT OUTER | | 55M| 10G| | 21643 (2)| 00:04:20 |
| 7 | TABLE ACCESS FULL | TB_USER_AGENT_RELAT | 27937 | 1200K| | 102 (0)| 00:00:02 |
|* 8 | HASH JOIN OUTER | | 3374K| 511M| | 21392 (1)| 00:04:17 |
|* 9 | HASH JOIN SEMI | | 1712 | 188K| | 2007 (1)| 00:00:25 |
|* 10 | HASH JOIN RIGHT OUTER | | 1712 | 173K| | 32 (0)| 00:00:01 |
| 11 | TABLE ACCESS FULL | TB_USER_ZGY | 43 | 903 | | 3 (0)| 00:00:01 |
|* 12 | HASH JOIN RIGHT OUTER | | 1712 | 138K| | 29 (0)| 00:00:01 |
|* 13 | TABLE ACCESS FULL | OSS_USER_STATION | 1075 | 25800 | | 6 (0)| 00:00:01 |
| 14 | TABLE ACCESS FULL | TB_AGENT_INFO | 1712 | 98K| | 23 (0)| 00:00:01 |
| 15 | VIEW | VW_NSO_1 | 16271 | 143K| | 1975 (1)| 00:00:24 |
|* 16 | VIEW | | 16271 | 143K| | 1975 (1)| 00:00:24 |
| 17 | HASH UNIQUE | | 16271 | 8882K| 10M| 1975 (1)| 00:00:24 |
|* 18 | CONNECT BY WITHOUT FILTERING (UNIQUE)| | | | | | |
|* 19 | HASH JOIN RIGHT SEMI | | 530 | 146K| | 29 (0)| 00:00:01 |
|* 20 | TABLE ACCESS FULL | TB_USER_CHANNEL | 600 | 7800 | | 7 (0)| 00:00:01 |
| 21 | TABLE ACCESS FULL | TB_CHANNEL_INFO | 1807 | 476K| | 22 (0)| 00:00:01 |
| 22 | TABLE ACCESS FULL | TB_CHANNEL_INFO | 1807 | 476K| | 22 (0)| 00:00:01 |
|* 23 | TABLE ACCESS FULL | BASE_DATA_INVEST_INFO | 3374K| 148M| | 19375 (1)| 00:03:53 |
----------------------------------------------------------------------------------------------------------------------------------
Predicate Information (identified by operation id):
---------------------------------------------------
6 - access("AGENT_ID"="C"."AGENT_ID"(+))
8 - access("AGENT_ID"="AGENT_ID"(+))
9 - access("AGENT_ID"="AGENT_ID")
10 - access("C"."USERNAME"="D"."USERNAME"(+))
12 - access("AGENT_ID"="C"."AGENT_ID"(+))
13 - filter("C"."USER_TYPE"(+)=0)
16 - filter("AGENT_ID" IS NOT NULL)
18 - access("T"."PARENT_CHANNEL_ID"=PRIOR "T"."CHANNEL_ID")
19 - access("T"."CHANNEL_ID"="CHANNEL_ID")
20 - filter("USER_ID"=596)
23 - filter("STR_DAY"(+)>='20150801' AND "STR_DAY"(+)<='20160821')

尝试单独跑 a,很快

1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
(select a.city,
a.agent_id,
a.username,
a.real_name, -- 业主姓名
a.phone, -- 业主手机号
d.real_name zgy_name, -- 所属专管员
count(distinct case
when c.str_day <= '20160821' then
c.login_name
end) login_count,
count(distinct case
when c.str_day <= '20160821' then
decode(c.status, 1, c.invest_id, null)
end) user_count
from (select agent_id, city, username, real_name, phone
from agent.TB_AGENT_INFO
where agent_id in
(SELECT agent_id
FROM (SELECT distinct *
FROM TB_CHANNEL_INFO t
START WITH t.CHANNEL_ID in
(select CHANNEL_ID
from TB_USER_CHANNEL
where USER_ID = 596)
CONNECT BY PRIOR
t.CHANNEL_ID = t.PARENT_CHANNEL_ID)
WHERE agent_id IS NOT NULL)) a
left join oss_user_station e
on a.agent_id = e.agent_id
and e.user_type = 0
left join tb_user_zgy d
on e.username = d.username
left join act.tb_user_agent_relat c
on a.agent_id = c.agent_id
group by a.city,
a.username,
a.real_name,
a.phone,
d.real_name,
a.agent_id) a

单独跑a很快,和b合在一起就很慢,那么怀疑是由于视图合并,导致了a内部的表提前去和b关联,引发了性能问题。
尝试禁止视图合并可以使用rownum>0,或no_merge hint

1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
61
62
63
64
65
66
67
68
69
70
71
72
73
74
75
76
77
78
79
80
81
82
83
84
85
86
87
88
89
90
91
92
93
94
95
96
97
98
99
100
101
102
103
104
105
106
107
108
109
110
select a.city,
a.agent_id,
a.username,
a.real_name,
phone,
zgy_name,
login_count,
user_count,
count(distinct b.invest_id) user_invested,
sum(b.order_amount / 100) invest_amount
from (select * from (select a.city,
a.agent_id,
a.username,
a.real_name, -- 业主姓名
a.phone, -- 业主手机号
d.real_name zgy_name, -- 所属专管员
count(distinct case
when c.str_day <= '20160821' then
c.login_name
end) login_count,
count(distinct case
when c.str_day <= '20160821' then
decode(c.status, 1, c.invest_id, null)
end) user_count
from (select agent_id, city, username, real_name, phone
from agent.TB_AGENT_INFO
where agent_id in
(SELECT agent_id
FROM (SELECT distinct *
FROM TB_CHANNEL_INFO t
START WITH t.CHANNEL_ID in
(select CHANNEL_ID
from TB_USER_CHANNEL
where USER_ID = 596)
CONNECT BY PRIOR
t.CHANNEL_ID = t.PARENT_CHANNEL_ID)
WHERE agent_id IS NOT NULL)) a
left join oss_user_station e
on a.agent_id = e.agent_id
and e.user_type = 0
left join tb_user_zgy d
on e.username = d.username
left join act.tb_user_agent_relat c
on a.agent_id = c.agent_id
group by a.city,
a.username,
a.real_name,
a.phone,
d.real_name,
a.agent_id) where rownum>0)a
left join (select invest_id, order_amount, agent_id, str_day
from agent.base_data_invest_info
where str_day >= '20150801' and str_day<='20160821') b
on a.agent_id = b.agent_id
group by a.city,
a.agent_id,
a.username,
a.real_name,
a.phone,
a.zgy_name,
a.login_count,
a.user_count
kuai
-----------------------------------------------------------------------------------------------------------------------------------
| Id | Operation | Name | Rows | Bytes |TempSpc| Cost (%CPU)| Time |
-----------------------------------------------------------------------------------------------------------------------------------
| 0 | SELECT STATEMENT | | 823M| 96G| | 23M (1)| 78:59:52 |
| 1 | HASH GROUP BY | | 823M| 96G| | 23M (1)| 78:59:52 |
| 2 | VIEW | VW_DAG_0 | 823M| 96G| | 23M (1)| 78:59:52 |
| 3 | HASH GROUP BY | | 823M| 98G| 112G| 23M (1)| 78:59:52 |
|* 4 | HASH JOIN OUTER | | 823M| 98G| 26M| 41358 (6)| 00:08:17 |
| 5 | VIEW | | 259K| 23M| | 11090 (1)| 00:02:14 |
| 6 | COUNT | | | | | | |
|* 7 | FILTER | | | | | | |
| 8 | VIEW | | 259K| 23M| | 11090 (1)| 00:02:14 |
| 9 | SORT GROUP BY | | 259K| 38M| 41M| 11090 (1)| 00:02:14 |
|* 10 | HASH JOIN | | 259K| 38M| | 2111 (1)| 00:00:26 |
|* 11 | VIEW | | 16271 | 143K| | 1975 (1)| 00:00:24 |
| 12 | HASH UNIQUE | | 16271 | 8882K| 10M| 1975 (1)| 00:00:24 |
|* 13 | CONNECT BY WITHOUT FILTERING (UNIQUE)| | | | | | |
|* 14 | HASH JOIN RIGHT SEMI | | 530 | 146K| | 29 (0)| 00:00:01 |
|* 15 | TABLE ACCESS FULL | TB_USER_CHANNEL | 600 | 7800 | | 7 (0)| 00:00:01 |
| 16 | TABLE ACCESS FULL | TB_CHANNEL_INFO | 1807 | 476K| | 22 (0)| 00:00:01 |
| 17 | TABLE ACCESS FULL | TB_CHANNEL_INFO | 1807 | 476K| | 22 (0)| 00:00:01 |
|* 18 | HASH JOIN OUTER | | 27937 | 4037K| | 134 (0)| 00:00:02 |
|* 19 | HASH JOIN RIGHT OUTER | | 1712 | 173K| | 32 (0)| 00:00:01 |
| 20 | TABLE ACCESS FULL | TB_USER_ZGY | 43 | 903 | | 3 (0)| 00:00:01 |
|* 21 | HASH JOIN RIGHT OUTER | | 1712 | 138K| | 29 (0)| 00:00:01 |
|* 22 | TABLE ACCESS FULL | OSS_USER_STATION | 1075 | 25800 | | 6 (0)| 00:00:01 |
| 23 | TABLE ACCESS FULL | TB_AGENT_INFO | 1712 | 98K| | 23 (0)| 00:00:01 |
| 24 | TABLE ACCESS FULL | TB_USER_AGENT_RELAT | 27937 | 1200K| | 102 (0)| 00:00:02 |
|* 25 | TABLE ACCESS FULL | BASE_DATA_INVEST_INFO | 3374K| 109M| | 19375 (1)| 00:03:53 |
-----------------------------------------------------------------------------------------------------------------------------------
Predicate Information (identified by operation id):
---------------------------------------------------
4 - access("A"."AGENT_ID"="AGENT_ID"(+))
7 - filter(ROWNUM>0)
10 - access("AGENT_ID"="AGENT_ID")
11 - filter("AGENT_ID" IS NOT NULL)
13 - access("T"."PARENT_CHANNEL_ID"=PRIOR "T"."CHANNEL_ID")
14 - access("T"."CHANNEL_ID"="CHANNEL_ID")
15 - filter("USER_ID"=596)
18 - access("AGENT_ID"="C"."AGENT_ID"(+))
19 - access("C"."USERNAME"="D"."USERNAME"(+))
21 - access("AGENT_ID"="C"."AGENT_ID"(+))
22 - filter("C"."USER_TYPE"(+)=0)
25 - filter("STR_DAY"(+)>='20150801' AND "STR_DAY"(+)<='20160821')

用no_merge hint禁止视图合并也可以

1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
61
62
63
select a.city,
a.agent_id,
a.username,
a.real_name,
phone,
zgy_name,
login_count,
user_count,
count(distinct b.invest_id) user_invested,
sum(b.order_amount / 100) invest_amount
from (select /*+ no_merge */
a.city,
a.agent_id,
a.username,
a.real_name, -- 业主姓名
a.phone, -- 业主手机号
d.real_name zgy_name, -- 所属专管员
count(distinct case
when c.str_day <= '20160821' then
c.login_name
end) login_count,
count(distinct case
when c.str_day <= '20160821' then
decode(c.status, 1, c.invest_id, null)
end) user_count
from (select /*+ qb_name(sb) */ agent_id, city, username, real_name, phone
from agent.TB_AGENT_INFO
where agent_id in
(SELECT agent_id
FROM (SELECT distinct *
FROM TB_CHANNEL_INFO t
START WITH t.CHANNEL_ID in
(select CHANNEL_ID
from TB_USER_CHANNEL
where USER_ID = 596)
CONNECT BY PRIOR
t.CHANNEL_ID = t.PARENT_CHANNEL_ID)
WHERE agent_id IS NOT NULL)) a
left join oss_user_station e
on a.agent_id = e.agent_id
and e.user_type = 0
left join tb_user_zgy d
on e.username = d.username
left join (select * from act.tb_user_agent_relat c) c
on a.agent_id = c.agent_id
group by a.city,
a.username,
a.real_name,
a.phone,
d.real_name,
a.agent_id) a
left join (select invest_id, order_amount, agent_id, str_day
from agent.base_data_invest_info
where str_day >= '20150801' and str_day<='20160821') b
on a.agent_id = b.agent_id
group by a.city,
a.agent_id,
a.username,
a.real_name,
a.phone,
a.zgy_name,
a.login_count,
a.user_count

至此sql从一个小时都跑不完,到最后两秒跑完,工作已经完成,但是单从慢的执行计划中并没有看出什么问题。有聚合函数group by走hash没有错,虽然有全表扫描带*但是要么过滤性太差,要么不是性能瓶颈。那为什么总共300多w就跑不完了呢

慢的执行计划做一个10046

1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
Number of plan statistics captured: 1
Rows (1st) Rows (avg) Rows (max) Row Source Operation
---------- ---------- ---------- ---------------------------------------------------
0 0 0 HASH GROUP BY (cr=0 pr=0 pw=0 time=278 us cost=3934270 size=6937507584 card=55059584)
0 0 0 VIEW VW_DAG_1 (cr=0 pr=0 pw=0 time=111 us cost=3934270 size=6937507584 card=55059584)
0 0 0 HASH GROUP BY (cr=0 pr=0 pw=0 time=108 us cost=3934270 size=6607150080 card=55059584)
0 0 0 VIEW VM_NWVW_0 (cr=0 pr=0 pw=0 time=32 us cost=2456206 size=6607150080 card=55059584)
0 0 0 SORT GROUP BY (cr=0 pr=0 pw=0 time=31 us cost=2456206 size=11177095552 card=55059584)
148234852 148234852 148234852 HASH JOIN RIGHT OUTER (cr=34882 pr=0 pw=0 time=34098445 us cost=21643 size=11177095552 card=55059584)
29651 29651 29651 TABLE ACCESS FULL TB_USER_AGENT_RELAT (cr=332 pr=0 pw=0 time=8201 us cost=102 size=1229228 card=27937)
703556 703556 703556 HASH JOIN OUTER (cr=34550 pr=0 pw=0 time=1518631 us cost=21392 size=536480628 card=3374092)
612 612 612 HASH JOIN SEMI (cr=272 pr=0 pw=0 time=31359 us cost=2007 size=193456 card=1712)
1751 1751 1751 HASH JOIN RIGHT OUTER (cr=100 pr=0 pw=0 time=11404 us cost=32 size=178048 card=1712)
43 43 43 TABLE ACCESS FULL TB_USER_ZGY (cr=2 pr=0 pw=0 time=103 us cost=3 size=903 card=43)
1751 1751 1751 HASH JOIN RIGHT OUTER (cr=98 pr=0 pw=0 time=6664 us cost=29 size=142096 card=1712)
1312 1312 1312 TABLE ACCESS FULL OSS_USER_STATION (cr=15 pr=0 pw=0 time=420 us cost=6 size=25800 card=1075)
1751 1751 1751 TABLE ACCESS FULL TB_AGENT_INFO (cr=83 pr=0 pw=0 time=1804 us cost=23 size=101008 card=1712)
612 612 612 VIEW VW_NSO_1 (cr=172 pr=0 pw=0 time=19720 us cost=1975 size=146439 card=16271)
612 612 612 VIEW (cr=172 pr=0 pw=0 time=19351 us cost=1975 size=146439 card=16271)
613 613 613 HASH UNIQUE (cr=172 pr=0 pw=0 time=19224 us cost=1975 size=9095489 card=16271)
1215 1215 1215 CONNECT BY WITHOUT FILTERING (UNIQUE) (cr=172 pr=0 pw=0 time=16687 us)
603 603 603 HASH JOIN RIGHT SEMI (cr=97 pr=0 pw=0 time=4922 us cost=29 size=149990 card=530)
603 603 603 TABLE ACCESS FULL TB_USER_CHANNEL (cr=22 pr=0 pw=0 time=550 us cost=7 size=7800 card=600)
1807 1807 1807 TABLE ACCESS FULL TB_CHANNEL_INFO (cr=75 pr=0 pw=0 time=1615 us cost=22 size=487890 card=1807)
1807 1807 1807 TABLE ACCESS FULL TB_CHANNEL_INFO (cr=75 pr=0 pw=0 time=1133 us cost=22 size=487890 card=1807)
1631878 1631878 1631878 TABLE ACCESS FULL BASE_DATA_INVEST_INFO (cr=34278 pr=0 pw=0 time=950767 us cost=19375 size=155208232 card=3374092)

id 6 1亿4千多万,一个多小时也没跑出来
并且temp撑爆了
第 43 行出现错误:
ORA-01652: 无法通过 128 (在表空间 TEMP 中) 扩展 temp 段
一亿四千多万,b表才300万,sql group by之前也不过一百多万的结果

根据 6 -
access(“AGENT_ID”=”C”.”AGENT_ID”(+)) 查看c和b表agent_id数据分布

1
select agent_id,count(*) from act.tb_user_agent_relat group by agent_id order by 2 desc

最多的6827行,最少的1行

1
select agent_id,count(*) from agent.base_data_invest_info group by agent_id order by 2 desc

最多50w,最少1行
又一次进了hash join链接列数据分布不均匀的坑,hash join只适合数据分布均匀的列做链接条件

做个oradebug short_stack

1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
SQL> select unique sid from v$mystat;
SID
----------
1132
SQL> select p.spid from v$process p ,v$session s where s.paddr=p.addr and s.sid=1132;
SPID
------------------------------------------------
28539
oradebug setospid 28539
SQL> oradebug short_stack
ksedsts()+465<-ksdxfstk()+32<-ksdxcb()+1927<-sspuser()+112<-__sighandler()<-io_submit()+7<-skgfqio()+1275<-ksfd_skgfqio()+894<-ksfdgo()+423<-ksfdaio()+2290<-kcflbi()+906<-kcbldio()+3104<-kcblsltio()+530<-stsIssueWrite()+118<-stsGetBlock()+442<-sdbinb()+135<-sdbput()+1042<-smbwrt()+247<-smbput()+2503<-sorput()+93<-qesaEvaAndPutDistAggOpns()+590<-qergsRowP()+430<-qerhjWalkHashBucket()+397<-qerhjGenProbeHashTable()+1571<-qerhjGenProbeHashTable()+718<-kdstf11011010000km()+673<-kdsttgr()+153241<-qertbFetch()+2455<-rwsfcd()+103<-qerhjFetch()+1661<-rwsfcd()+103<-qerhjFetch()+1661<-qergsFetch()+757<-qervwFetch()+139<-qerghFetch()+315<-qervwFetch()+139<-qerghFetch()+315<-opifch2()+2766<-kpoal8()+2833<-opiodr()+917<-ttcpip()+2183<-opitsk()+1710<-opiino()+969<-opiodr()+917<-opidrv()+570<-sou2o()+103<-opimai_real()+133<-ssthrdmain()+265<-main()+201<-__libc_start_main()+244
SQL>
SQL>
SQL>
SQL>
SQL> oradebug short_stack
ksedsts()+465<-ksdxfstk()+32<-ksdxcb()+1927<-sspuser()+112<-__sighandler()<-io_submit()+7<-skgfqio()+1275<-ksfd_skgfqio()+894<-ksfdgo()+423<-ksfdaio()+2290<-kcflbi()+906<-kcbldio()+3104<-kcblsltio()+530<-stsIssueWrite()+118<-stsGetBlock()+442<-sdbinb()+135<-sdbput()+1042<-smbwrt()+247<-smbput()+2503<-sorput()+93<-qesaEvaAndPutDistAggOpns()+590<-qergsRowP()+430<-qerhjWalkHashBucket()+397<-qerhjGenProbeHashTable()+1571<-qerhjWalkHashBucket()+397<-qerhjGenProbeHashTable()+1571<-kdstf11011010000km()+673<-kdsttgr()+153241<-qertbFetch()+2455<-rwsfcd()+103<-qerhjFetch()+1661<-rwsfcd()+103<-qerhjFetch()+1661<-qergsFetch()+757<-qervwFetch()+139<-qerghFetch()+315<-qervwFetch()+139<-qerghFetch()+315<-opifch2()+2766<-kpoal8()+2833<-opiodr()+917<-ttcpip()+2183<-opitsk()+1710<-opiino()+969<-opiodr()+917<-opidrv()+570<-sou2o()+103<-opimai_real()+133<-ssthrdmain()+265<-main()+201<-__libc_start_main()+244
SQL>
SQL>
SQL>
SQL>
SQL> oradebug short_stack
ksedsts()+465<-ksdxfstk()+32<-ksdxcb()+1927<-sspuser()+112<-__sighandler()<-qergsRowP()+2161<-qerhjWalkHashBucket()+397<-qerhjGenProbeHashTable()+1571<-qerhjWalkHashBucket()+397<-qerhjGenProbeHashTable()+1571<-kdstf11011010000km()+673<-kdsttgr()+153241<-qertbFetch()+2455<-rwsfcd()+103<-qerhjFetch()+1661<-rwsfcd()+103<-qerhjFetch()+1661<-qergsFetch()+757<-qervwFetch()+139<-qerghFetch()+315<-qervwFetch()+139<-qerghFetch()+315<-opifch2()+2766<-kpoal8()+2833<-opiodr()+917<-ttcpip()+2183<-opitsk()+1710<-opiino()+969<-opiodr()+917<-opidrv()+570<-sou2o()+103<-opimai_real()+133<-ssthrdmain()+265<-main()+201<-__libc_start_main()+244
SQL> oradebug short_stack
ksedsts()+465<-ksdxfstk()+32<-ksdxcb()+1927<-sspuser()+112<-__sighandler()<-lmebco()+63<-qesaSimpleCompare()+73<-smbput()+913<-sorput()+93<-qergsRowP()+1067<-qerhjWalkHashBucket()+397<-qerhjGenProbeHashTable()+1571<-qerhjGenProbeHashTable()+718<-kdstf11011010000km()+673<-kdsttgr()+153241<-qertbFetch()+2455<-rwsfcd()+103<-qerhjFetch()+1661<-rwsfcd()+103<-qerhjFetch()+1661<-qergsFetch()+757<-qervwFetch()+139<-qerghFetch()+315<-qervwFetch()+139<-qerghFetch()+315<-opifch2()+2766<-kpoal8()+2833<-opiodr()+917<-ttcpip()+2183<-opitsk()+1710<-opiino()+969<-opiodr()+917<-opidrv()+570<-sou2o()+103<-opimai_real()+133<-ssthrdmain()+265<-main()+201<-__libc_start_main()+244
可以看到qerhjWalkHashBucket
qerhjWalkHashBucket就表示在做hash join的过程中需要遍历hash bucket中的数据,当链接列数据分布不均,某些值特别多时,遍历其hash bucket的成本也就非常高,如果pga放不下了,就会放到temp进行磁盘io,这就是性能瓶颈的原因,这个例子把30g的temp表空间都撑爆了,可见hash bucket有多大!

做个SQL MONITOR,也可以看出,瓶颈在id 6。如果做一个sql rpt也可以发现sql执行过程中的每妙逻辑读实际并不高,因为时间都花在了遍历hash bucket中

Powered by Hexo and Hexo-theme-hiker

Copyright © 2013 - 2018 Fan() All Rights Reserved.

访客数 : | 访问量 :

#