mysql性能优化-explain

一、id:select查询的序号，包含一组数字，表示查询中执行select字句或者操作顺序。
1、id相同，自上到下顺序执行；
2、id不同，序号大的先执行；
3、id相同不同共存，序号大的先执行，相同的顺序执行；
二、select_type: 查询类型，主要用于区别普通查询、联合查询、子查询等复杂查询
常见simple、primary、subquery、derived、union、union result六种
1、simple：简单的select查询，查询中不包含子查询或者union；
2、primary：查询中若包含任何复杂的子部分，最外层查询则被标记为primary；
3、subquery：在select或者where列表中包含子查询；
4、derived：在from列表中包含的子查询被标记为derived，
mysql会递归执行这些子查询，把结果放在临时表中，临时表会增加系统负担；
5、union：若第二个select出现在union之后，则被标记为union；
若union包含在from字句的查询中，外层select将被标记为：derived；
6、union result：从union表获取的select；
三、table：这行数据关于哪张表。
四、type：
system > const > eq_ref > ref > range > index > all
一般来说，得保证查询至少达到range级别，最好能达到ref
system 表只有一行记录（系统表）
const 表示通过索引一次就可以找到了，const用于比较primary key或者unique列索引。因为只匹配一行数据，所以很快
如：将primary key放置wehere中
eq_ref 唯一性索引扫描。对于每个索引键，表中只有一条记录与之匹配，常见于主键或者唯一索引扫描
explain select au.user_name, an.unit_name  from acc_user au left join acc_unit an on au.unit_id = an.unit_id；
+----+-------------+-------+------------+--------+---------------+---------+---------+--------------------------+------+----------+-------+
| id | select_type | table | partitions | type   | possible_keys | key     | key_len | ref                      | rows | filtered | Extra |
+----+-------------+-------+------------+--------+---------------+---------+---------+--------------------------+------+----------+-------+
|  1 | SIMPLE      | au    | NULL       | ALL    | NULL          | NULL    | NULL    | NULL                     |   28 |   100.00 | NULL  |
|  1 | SIMPLE      | an    | NULL       | eq_ref | PRIMARY       | PRIMARY | 8       | smartpatroldb.au.unit_id |    1 |   100.00 | NULL  |
+----+-------------+-------+------------+--------+---------------+---------+---------+--------------------------+------+----------+-------+
ref 非唯一性索引扫描，返回匹配某个单独值的所有行
explain select au.user_name, an.unit_name  from acc_unit an left join acc_user au on an.unit_id = au.unit_id;
+----+-------------+-------+------------+-------+------------------------+------------------------+---------+--------------------------+------+----------+-------------+
| id | select_type | table | partitions | type  | possible_keys          | key                    | key_len | ref                      | rows | filtered | Extra       |
+----+-------------+-------+------------+-------+------------------------+------------------------+---------+--------------------------+------+----------+-------------+
|  1 | SIMPLE      | an    | NULL       | index | NULL                   | idx_unit_name          | 767     | NULL                     |    6 |   100.00 | Using index |
|  1 | SIMPLE      | au    | NULL       | ref   | FK_Reference_unit_user | FK_Reference_unit_user | 9       | smartpatroldb.an.unit_id |    7 |   100.00 | NULL        |
+----+-------------+-------+------------+-------+------------------------+------------------------+---------+--------------------------+------+----------+-------------+
range 索引给定范围行，使用一个索引选择行
explain select * from acc_user where user_id between 1000 and 3000;
+----+-------------+----------+------------+-------+---------------+---------+---------+------+------+----------+-------------+
| id | select_type | table    | partitions | type  | possible_keys | key     | key_len | ref  | rows | filtered | Extra       |
+----+-------------+----------+------------+-------+---------------+---------+---------+------+------+----------+-------------+
|  1 | SIMPLE      | acc_user | NULL       | range | PRIMARY       | PRIMARY | 8       | NULL |    2 |   100.00 | Using where |
+----+-------------+----------+------------+-------+---------------+---------+---------+------+------+----------+-------------+
index Full Index Scan，index与all区别，index只扫描索引树。通常比all快。
all 全表扫描
五、possible_keys: 显示可能应用到这张表的索引，一个或多个；但不一定查询实际使用到；
六、key：使用到索引；
explain select * from p_base_data where patrol_task_id = 1；
+----+-------------+-------------+------------+-------+------------------------------+---------------------------------+---------+------+------+----------+-------------+
| id | select_type | table       | partitions | type  | possible_keys                |                         key     | key_len | ref  | rows | filtered | Extra       |
+----+-------------+-------------+------------+-------+------------------------------+---------------------------------+---------+------+------+----------+-------------+
|  1 | SIMPLE      | p_base_data | NULL       | ref   | idx_p_base_data_patrolTaskId | idx_p_base_data_patrolTaskId    | 9       | const|    2 |   100.00 | Using where |
+----+-------------+-------------+------------+-------+------------------------------+---------------------------------+---------+------+------+----------+-------------+	
七、key_len：索引中使用的字节个数，可以通过该列计算查询中索引使用的长度；
explain select * from t1 where col1='ab';
+----+-------------+----------+------------+-------+---------------+---------------+---------+------+------+----------+-------------+
| id | select_type | table    | partitions | type  | possible_keys | key           | key_len | ref  | rows | filtered | Extra       |
+----+-------------+----------+------------+-------+---------------+---------------+---------+------+------+----------+-------------+
|  1 | SIMPLE      | t1       | NULL       | ref   | idx_col1_col2 | idx_col1_col2 | 13      | NULL |  143 |   100.00 | Using where |
+----+-------------+----------+------------+-------+---------------+---------------+---------+------+------+----------+-------------+
explain select * from t1 where col1='ab' and col2='ac';
+----+-------------+----------+------------+-------+---------------+---------------+---------+------+------+----------+-------------+
| id | select_type | table    | partitions | type  | possible_keys | key           | key_len | ref  | rows | filtered | Extra       |
+----+-------------+----------+------------+-------+---------------+---------------+---------+------+------+----------+-------------+
|  1 | SIMPLE      | t1       | NULL       | ref   | idx_col1_col2 | idx_col1_col2 | 26      | NULL |   1  |   100.00 | Using where |
+----+-------------+----------+------------+-------+---------------+---------------+---------+------+------+----------+-------------+
八、ref：显示索引列哪一列被使用了，如果可能的话，是一个常量；
九、rows: 根据表统计信息及索引使用情况。大致估算出找出所需读取行数； 越小越好
十、Extra：
using filesort
using temporary
using index
using where
using join buffer