PostgreSQL技术大讲堂 - 第30讲：多表连接方式

admin

PostgreSQL从小白到专家，是从入门逐渐能力提升的一个系列教程，内容包括对PG基础的认知、包括安装使用、包括角色权限、包括维护管理、、等内容，希望对热爱PG、学习PG的同学们有帮助，欢迎持续关注CUUG PG技术大讲堂。

第30讲：多表连接方式

第30讲预告：9月23日(周六)19:30-20:30，钉钉群直播，群号：35822460

内容2 : Merge Join连接方式

多表连接方式

nested loop join

hash join

NATURAL INNER JOIN

LEFT/RIGHT OUTER JOIN

Nested Loop Join

Nested Loop Join图解

testdb=# EXPLAIN SELECT * FROM tbl_a AS a, tbl_b AS b WHERE a.id = b.id;

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

Join Filter: (a.id = b.id)

-> Materialize (cost=0.00..98.00 rows=5000 width=8)

(5 rows)

Materialize成本估算

(Materialized) Nested Loop成本估算

QUERY PLAN

Nested Loop (cost=0.29..1935.50 rows=5000 width=16)

-> Index Scan using tbl_c_pkey on tbl_c c (cost=0.29..0.36 rows=1 width=8)

（4 rows）

具有外部索引扫描的嵌套循环联接的三种变体

Merge Join连接方式

Merge Join成本估算

testdb=# EXPLAIN SELECT * FROM tbl_a AS a, tbl_b AS b WHERE a.id = b.id AND b.id < 1000;

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

Merge Cond: (a.id = b.id)

Sort Key: a.id

-> Sort (cost=135.33..137.83 rows=1000 width=8)

-> Seq Scan on tbl_b b (cost=0.00..85.50 rows=1000 width=8)

(9 rows)

Materialized Merge Join

testdb=# SET enable_hashjoin TO off;

testdb=# EXPLAIN SELECT * FROM tbl_c AS c, tbl_b AS b WHERE c.id = b.id AND b.id < 1000;

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

Merge Cond: (c.id = b.id)

-> Sort (cost=135.33..137.83 rows=1000 width=8)

-> Seq Scan on tbl_b b (cost=0.00..85.50 rows=1000 width=8)

(7 rows)

materialized merge join with outer index scan

testdb=# SET enable_nestloop TO off;

QUERY PLAN

Merge Join (cost=421.84..672.09 rows=4500 width=16)

-> Index Scan using tbl_c_pkey on tbl_c c (cost=0.29..318.29 rows=10000 width=8)

-> Sort (cost=421.55..432.80 rows=4500 width=8)

-> Seq Scan on tbl_b b (cost=0.00..85.50 rows=4500 width=8)

(8 rows)

indexed merge join with outer index scan

testdb=# SET enable_nestloop TO off;

QUERY PLAN

Merge Join (cost=0.57..226.07 rows=1000 width=16)

-> Index Scan using tbl_c_pkey on tbl_c c (cost=0.29..318.29 rows=10000 width=8)

Index Cond: (id < 1000)

Hash Join

In-Memory Hash Join

将内部表的所有元组插入到一个批处理中

将外部表的每个元组与批处理中的内部元组进行比较，如果满足连接条件，则进行连接

Hash Join

预处理

2、向上拉子查询

testdb=# SELECT * FROM tbl_a AS a, (SELECT * FROM tbl_b) as b WHERE a.id = b.id;

3、将外部联接转换为内部联接

优化器可用规则

1、表数量小于12张，应用动态规划得到最优的计划

参数 geqo_threshold指定的阈值(默认值为12)

SGetting the Cheapest Path of a Triple-Table Query

testdb=# WHERE a.id = b.id AND b.id = c.id AND a.data < 40;

{tbl_a,tbl_b,tbl_c}=min({tbl_a,{tbl_b,tbl_c}},{tbl_b,{tbl_a,tbl_c}},{tbl_c,{tbl_a,tbl_b}}).

以上就是第30讲的内容，往期视频及文档，联系CUUG