允许分布在不同Node Group中的表进行关联查询。包括基表扫描操作、关联操作、集合操作、聚集操作和增删改操作。

用户可通过GUC参数对计算Node Group进行调整，具体调整方式请参考expected_computing_nodegroup。

示例

示例一：多Node Group关联使用

建立Node Group和基表

create node group ng1 with (dn_6001_6002, dn_6003_6004);
create node group ng2 with (dn_6005_6006, dn_6007_6008, dn_6009_6010);
create table t1 (a int, b int, c int, d int) to group ng1;
create table t2 (a int, b int, c int, d int) to group ng2;

查看Node Group

select group_name from PGXC_GROUP;

增删改操作

insert into t1 select * from t2;
delete from t1 using t2 where t1.b = t2.c;
update t1 set b = 1 from t2 where t1.c = t2.c;

基表扫描操作
```
select b from t1 where c=6;
```
```
关联操作
```

select * from t1, t2 where t1.b = t2.c;

集合操作

select * from t1 union all select * from t2;
select * from t1 intersect select * from t2;

聚集操作

select b, sum(c) from t1 group by b;
select t1.b, t2.b, avg(t1.c + t2.c) from t1, t2 where t1.d = t2.d group by t1.b, t2.b;

ITPub博客

使用Node Group-多Node Group关联查询

示例

最新文章