mysql-联接查询-增加索引

联接查询-增加索引（对于内连接）

更多情况：联接查询-驱动表

1.对应的表

CREATE TABLE `dept_emp` (
`emp_no` int(11) NOT NULL,
`dept_no` char(4) NOT NULL,
`from_date` date NOT NULL,
`to_date` date NOT NULL,
PRIMARY KEY (`emp_no`,`dept_no`));
CREATE TABLE `salaries` (
`emp_no` int(11) NOT NULL,
`salary` int(11) NOT NULL,
`from_date` date NOT NULL,
`to_date` date NOT NULL,
PRIMARY KEY (`emp_no`,`from_date`));

2.连接查询（不带索引）

EXPLAIN select *   from   dept_emp as d inner join salaries  as s  on s.to_date=d.to_date;
+----+-------------+-------+------------+------+---------------+------+---------+------+------+----------+----------------------------------------------------+
| id | select_type | table | partitions | type | possible_keys | key  | key_len | ref  | rows | filtered | Extra                                              |
+----+-------------+-------+------------+------+---------------+------+---------+------+------+----------+----------------------------------------------------+
|  1 | SIMPLE      | d     | NULL       | ALL  | NULL          | NULL | NULL    | NULL |  110 |   100.00 | NULL                                               |
|  1 | SIMPLE      | s     | NULL       | ALL  | NULL          | NULL | NULL    | NULL | 2814 |    10.00 | Using where; Using join buffer (Block Nested Loop) |
+----+-------------+-------+------------+------+---------------+------+---------+------+------+----------+----------------------------------------------------+

显然对于没有加索引的两张表，两张表全部进行的全表查询，查询次数1102814，很慢*

2.连接查询（数据少的表少上索引）

CREATE index to_date_ids on dept_emp(to_date)
EXPLAIN select *   from   dept_emp as d inner join salaries  as s  on s.to_date=d.to_date;
+----+-------------+-------+------------+------+---------------+-------------+---------+----------------+------+----------+-------+
| id | select_type | table | partitions | type | possible_keys | key         | key_len | ref            | rows | filtered | Extra |
+----+-------------+-------+------------+------+---------------+-------------+---------+----------------+------+----------+-------+
|  1 | SIMPLE      | s     | NULL       | ALL  | NULL          | NULL        | NULL    | NULL           | 2814 |   100.00 | NULL  |
|  1 | SIMPLE      | d     | NULL       | ref  | to_date_ids   | to_date_ids | 3       | test.s.to_date |   36 |   100.00 | NULL  |
+----+-------------+-------+------------+------+---------------+-------------+---------+----------------+------+----------+-------+

表dept_emp加上了索引，查询次数减少为362814*

3.连接查询（数据多的表加上索引）

drop index to_date_ids  on dept_emp
CREATE index to_date_ids on salaries(to_date)
EXPLAIN select *   from   dept_emp as d inner join salaries  as s  on s.to_date=d.to_date;
+----+-------------+-------+------------+------+---------------+-------------+---------+----------------+------+----------+-------+
| id | select_type | table | partitions | type | possible_keys | key         | key_len | ref            | rows | filtered | Extra |
+----+-------------+-------+------------+------+---------------+-------------+---------+----------------+------+----------+-------+
|  1 | SIMPLE      | d     | NULL       | ALL  | NULL          | NULL        | NULL    | NULL           |  110 |   100.00 | NULL  |
|  1 | SIMPLE      | s     | NULL       | ref  | to_date_ids   | to_date_ids | 3       | test.d.to_date |   33 |   100.00 | NULL  |
+----+-------------+-------+------------+------+---------------+-------------+---------+----------------+------+----------+-------+
CREATE index to_date_ids on dept_emp(to_date)

对于数据多的表salaries使用到了索引，只查询了33条数据，而表 dept_emp进行了全表查询，查询次数减少为110*33，极大的优化了查询次数

4.连接查询（都加上索引）

CREATE index to_date_ids on dept_emp(to_date)
EXPLAIN select *   from   dept_emp as d inner join salaries  as s  on s.to_date=d.to_date;
+----+-------------+-------+------------+------+---------------+-------------+---------+----------------+------+----------+-------+
| id | select_type | table | partitions | type | possible_keys | key         | key_len | ref            | rows | filtered | Extra |
+----+-------------+-------+------------+------+---------------+-------------+---------+----------------+------+----------+-------+
|  1 | SIMPLE      | d     | NULL       | ALL  | NULL          | NULL        | NULL    | NULL           |  110 |   100.00 | NULL  |
|  1 | SIMPLE      | s     | NULL       | ref  | to_date_ids   | to_date_ids | 3       | test.d.to_date |   33 |   100.00 | NULL  |
+----+-------------+-------+------------+------+---------------+-------------+---------+----------------+------+----------+-------+

结果同上

结论：

1.当联接查询两张表时，某一张表A存在索引，那么另一张表B会进行全表查询。在表B中查出一条记录，再去表A中查询是根据索引查询是否满足联接条件，满足则联接成功。
2.对于两张表都加上索引，对于上一种情况（数据多的表加索引）并不会有优化
所以应该选择记录多的表加上索引，避免数据多的表进行全表查询
就是将记录多的表作为被驱动表（对于内连接，驱动表选择方式联接查询-驱动表），对其加上索引。
ps：选择加索引时，数据量的多少，是不重复的数据量