postgresql-删除重复数据

时间:2023-01-30 11:40:08
greenplum最终的方法是:
delete from test where (gp_segment_id, ctid) not in (select gp_segment_id, min(ctid) from test group by x, gp_segment_id);
postgresql :
delete from t_zhzx_aj where ctid not in (select min(ctid) from db_zxaj.t_zhzx_aj group by ahdm)
使用窗口函数删除重复数据:
https://www.cnblogs.com/mchina/archive/2013/04/15/3022086.html
 PG数据库几种删除重复数据的方法
在使用PG数据库的这段时间,总结了三种删除重复数据的方法,其中最容易想到的就是最常规的删除方法,但此方法性能较差,删数据耗时较久,虽容易实现,但性能太差,影响写数据的速率。

另外就是被使用的group by删除方法,效率较高。

还有一种是刚发现的,还没有验证,现在就总结下这三种删除方法,并验证各自的执行效率。

首先创建一张基础表,并插入一定量的重复数据。

test=# create table deltest(id int, name varchar(255));
    CREATE TABLE
    test=# create table deltest_bk (like deltest);
    CREATE TABLE
    test=# insert into deltest select generate_series(1, 10000), 'ZhangSan';
    INSERT 0 10000
    test=# insert into deltest select generate_series(1, 10000), 'ZhangSan';
    INSERT 0 10000
    test=# insert into deltest_bk select * from deltest;
1
2
3
4
5
6
7
8
9
常规删除方法
最容易想到的方法就是判断数据是否重复,对于重复的数据只保留ctid最小(或最大)的那条数据,删除其他的数据。

test=# explain analyse delete from deltest a where a.ctid <> (select min(t.ctid) from deltest t where a.id=t.id);
                                                             QUERY PLAN
    -----------------------------------------------------------------------------------------------------------------------------
    Delete on deltest a  (cost=0.00..195616.30 rows=1518 width=6) (actual time=67758.866..67758.866 rows=0 loops=1)
       ->  Seq Scan on deltest a  (cost=0.00..195616.30 rows=1518 width=6) (actual time=32896.517..67663.228 rows=10000 loops=1)
         Filter: (ctid <> (SubPlan 1))
         Rows Removed by Filter: 10000
         SubPlan 1
           ->  Aggregate  (cost=128.10..128.10 rows=1 width=6) (actual time=3.374..3.374 rows=1 loops=20000)
                 ->  Seq Scan on deltest t  (cost=0.00..128.07 rows=8 width=6) (actual time=0.831..3.344 rows=2 loops=20000)
                       Filter: (a.id = id)
                       Rows Removed by Filter: 19998
    Total runtime: 67758.931 ms
    test=# select count(*) from deltest;
    count
    -------
    10000
    (1 行记录)
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
可以看到,id相同的数据,保留ctid最小的那条,其他的删除。相当于把deltest表中的数据删掉一半,耗时达到67s多。相当慢。

group by删除方法
第二种方法为group by方法,通过分组找到ctid最小的数据,然后删除其他数据。

test=# truncate table deltest;
    TRUNCATE TABLE
    test=# insert into deltest select * from deltest_bk;
    INSERT 0 20000
    test=# explain analyse delete from deltest a where a.ctid not in (select min(ctid) from deltest group by id);
                                                             QUERY PLAN
    ----------------------------------------------------------------------------------------------------------------------------------
    Delete on deltest a  (cost=131.89..2930.46 rows=763 width=6) (actual time=30942.496..30942.496 rows=0 loops=1)
       ->  Seq Scan on deltest a  (cost=131.89..2930.46 rows=763 width=6) (actual time=10186.296..30814.366 rows=10000 loops=1)
         Filter: (NOT (SubPlan 1))
         Rows Removed by Filter: 10000
         SubPlan 1
           ->  Materialize  (cost=131.89..134.89 rows=200 width=10) (actual time=0.001..0.471 rows=7500 loops=20000)
                 ->  HashAggregate  (cost=131.89..133.89 rows=200 width=10) (actual time=10.568..13.584 rows=10000 loops=1)
                       ->  Seq Scan on deltest  (cost=0.00..124.26 rows=1526 width=10) (actual time=0.006..3.829 rows=20000 loops=1)
     Total runtime: 30942.819 ms
    (9 行记录)
    test=# select count(*) from deltest;
     count
    -------
    10000
    (1 行记录)
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
可以看到同样是删除一半的数据,使用group by的方式,时间节省了一半。但仍含需要30s,下面试一下第三种删除操作。

新的删除方法
在postgres修炼之道这本书中,作者提到一种效率较高的删除方法, 在这里验证一下,具体如下:

test=# truncate table deltest;
    TRUNCATE TABLE
    test=# insert into deltest select * from deltest_bk;
    INSERT 0 20000                                                         
    test=# explain analyze delete from deltest a where a.ctid = any(array (select ctid from (select row_number() over (partition by id), ctid from deltest) t where t.row_number > 1));
                                                             QUERY PLAN
    ----------------------------------------------------------------------------------------------------------------------------------
    Delete on deltest a  (cost=250.74..270.84 rows=10 width=6) (actual time=98.363..98.363 rows=0 loops=1)
    InitPlan 1 (returns $0)
     ->  Subquery Scan on t  (cost=204.95..250.73 rows=509 width=6) (actual time=29.446..47.867 rows=10000 loops=1)
           Filter: (t.row_number > 1)
           Rows Removed by Filter: 10000
           ->  WindowAgg  (cost=204.95..231.66 rows=1526 width=10) (actual time=29.436..44.790 rows=20000 loops=1)
                 ->  Sort  (cost=204.95..208.77 rows=1526 width=10) (actual time=12.466..13.754 rows=20000 loops=1)
                       Sort Key: deltest.id
                       Sort Method: quicksort  Memory: 1294kB
                       ->  Seq Scan on deltest  (cost=0.00..124.26 rows=1526 width=10) (actual time=0.021..5.110 rows=20000 loops=1)
    ->  Tid Scan on deltest a  (cost=0.01..20.11 rows=10 width=6) (actual time=82.983..88.751 rows=10000 loops=1)
         TID Cond: (ctid = ANY ($0))
    Total runtime: 98.912 ms
    (13 行记录)

test=# select count(*) from deltest;
    count
    -------
    10000
    (1 行记录)
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
看到上述结果,真让我吃惊了一把,这么快的删除方法还是首次看到,自己真实孤陋寡闻,在这里要膜拜一下修炼之道这本书的大神作者了。
同时,努力向大神学习!!
---------------------
作者:arcticJian
来源:CSDN
原文:https://blog.csdn.net/arcticJian/article/details/50042647
版权声明:本文为博主原创文章,转载请附上博文链接!