MySQL执行计划核心内容
大家都知道，mysql在执行查询的时候会进行查询优化。简单来讲就是执行的时候先基于成本和规则优化生成执行计划，然后再按照执行计划执行查询。本文主要介绍EXPLAIN各输出项的含义，从而帮助大家更好的进行sql性能优化！
案例表设计
-- 用户表
create table t_user (
id int primary key,
login_name varchar(100),
name varchar(100),
age int,
sex char(1),
department_id int,
address varchar(100)
);
-- 部门表
create table t_department (
id int primary key,
name varchar(100)
);
-- 地址表
create table t_address (
id int primary key,
addr varchar(100)
);
-- 创建普通索引
alter table t_user
add index idx_dep(department_id);
-- 创建唯一索引
alter table t_user
add unique index idx_login_name(login_name);
alter table t_user
add index idx_name_age_sex(name, age, sex); -- 创建组合索引
alter table t_address
add fulltext ft_address(addr);-- 创建全文索引
执行计划详解
我们可以在查询语句前面加上EXPLAIN关键字来查看这个查询的执行计划。例如
mysql> EXPLAIN SELECT 1;
+----+-------------+-------+------------+------+---------------+------+---------+------+------+----------+----------------+| id | select_type | table | partitions | type | possible_keys | key | key_len | ref | rows | filtered | Extra |+----+-------------+-------+------------+------+---------------+------+---------+------+------+----------+----------------+| 1 | SIMPLE | NULL | NULL | NULL | NULL | NULL | NULL | NULL | NULL | NULL | No tables used |+----+-------------+-------+------------+------+---------------+------+---------+------+------+----------+----------------+1 row in set, 1 warning (0.01 sec)
可以看到，执行计划包含很多输出列，我们先简单过一下各列的大致作用，后面再进行详细讲解。
| 列名 | 描述 |
|---|---|
| id | 在一个大的查询语句中每个SELECT关键字都对应一个唯一的id |
| select_type | SELECT关键字对应的那个查询的类型 |
| table | 表名 |
| partitions | 匹配的分区信息 |
| type | 针对单表的访问方法 |
| possible_keys | 可能用到的索引 |
| key | 实际上使用的索引 |
| key_len | 实际使用到的索引长度 |
| ref | 当使用索引列等值查询时，与索引列进行等值匹配的对象信息 |
| rows | 预估的需要读取的记录条数 |
| filtered | 某个表经过搜索条件过滤后剩余记录条数的百分比 |
| Extra | 一些额外的信息 |
列详解
id
每个 SELECT语句都会自动分配的一个唯一标识符.
表示查询中操作表的顺序，有三种情况:
id相同:执行顺序由上到下 id不同:如果是子查询，id号会自增，id越大，优先级越高。 id相同的不同的同时存在
id列为null的就表示这是一个结果集，不需要使用它来进行查询。
select_type(重要)
查询类型，主要用于区别普通查询、联合查询(union、union all)、子查询等复杂查询。
simple
表示不需要union操作或者不包含子查询的简单select查询。有连接查询时，外层的查询为simple，且只有一个
mysql> explain select * from t_user;
+----+-------------+--------+------+---------------+------+---------+------+------+-------+
| id | select_type | table | type | possible_keys | key | key_len | ref | rows | Extra |
+----+-------------+--------+------+---------------+------+---------+------+------+-------+
| 1 | SIMPLE | t_user | ALL | NULL | NULL | NULL | NULL | 1 | NULL |
+----+-------------+--------+------+---------------+------+---------+------+------+-------+
1 row in set (0.00 sec)
primary
一个需要union操作或者含有子查询的select，位于最外层的单位查询的select_type即为primary。且只有一个
mysql> explain select (select name from t_user) from t_user ;
+----+-------------+--------+-------+---------------+---------+---------+------+------+-------------+
| id | select_type | table | type | possible_keys | key | key_len | ref | rows | Extra |
+----+-------------+--------+-------+---------------+---------+---------+------+------+-------------+
| 1 | PRIMARY | t_user | index | NULL | idx_dep | 5 | NULL | 1 | Using index |
| 2 | SUBQUERY | t_user | ALL | NULL | NULL | NULL | NULL | 1 | NULL |
+----+-------------+--------+-------+---------------+---------+---------+------+------+-------------+
2 rows in set (0.00 sec)
subquery
除了from字句中包含的子查询外，其他地方出现的子查询都可能是subquery
mysql> explain select * from t_user where id = (select max(id) from t_user);
+----+-------------+-------+------+---------------+------+---------+------+------+-----------------------------------------------------+
| id | select_type | table | type | possible_keys | key | key_len | ref | rows | Extra |
+----+-------------+-------+------+---------------+------+---------+------+------+-----------------------------------------------------+
| 1 | PRIMARY | NULL | NULL | NULL | NULL | NULL | NULL | NULL | Impossible WHERE noticed after reading const tables |
| 2 | SUBQUERY | NULL | NULL | NULL | NULL | NULL | NULL | NULL | No matching min/max row |
+----+-------------+-------+------+---------------+------+---------+------+------+-----------------------------------------------------+
2 rows in set (0.01 sec)
dependent subquery
与dependent union类似，表示这个subquery的查询要受到外部表查询的影响
mysql> explain select id,name,(select name from t_department a where a.id=b.department_id) from t_user b;
+----+--------------------+-------+--------+---------------+---------+---------+------------------------+------+-------+
| id | select_type | table | type | possible_keys | key | key_len | ref | rows | Extra |
+----+--------------------+-------+--------+---------------+---------+---------+------------------------+------+-------+
| 1 | PRIMARY | b | ALL | NULL | NULL | NULL | NULL | 1 | NULL |
| 2 | DEPENDENT SUBQUERY | a | eq_ref | PRIMARY | PRIMARY | 4 | sample.b.department_id | 1 | NULL |
+----+--------------------+-------+--------+---------------+---------+---------+------------------------+------+-------+
union
union连接的两个select查询，第一个查询是PRIMARY，除了第一个表外，第二个以后的表select_type 都是union
mysql> explain select * from t_user where sex='1' union select * from t_user
-> where sex='2';
+----+--------------+------------+------+---------------+------+---------+------+------+-----------------+
| id | select_type | table | type | possible_keys | key | key_len | ref | rows | Extra |
+----+--------------+------------+------+---------------+------+---------+------+------+-----------------+
| 1 | PRIMARY | t_user | ALL | NULL | NULL | NULL | NULL | 1 | Using where |
| 2 | UNION | t_user | ALL | NULL | NULL | NULL | NULL | 1 | Using where |
| NULL | UNION RESULT | <union1,2> | ALL | NULL | NULL | NULL | NULL | NULL | Using temporary |
+----+--------------+------------+------+---------------+------+---------+------+------+-----------------+
dependent union
与union一样，出现在union 或union all语句中，但是这个查询要受到外部查询的影响
mysql> explain select * from t_user where sex in (select sex from t_user where sex='1' union select sex from t_user where sex='2');
+----+--------------------+------------+------+---------------+------+---------+------+------+-----------------+
| id | select_type | table | type | possible_keys | key | key_len | ref | rows | Extra |
+----+--------------------+------------+------+---------------+------+---------+------+------+-----------------+
| 1 | PRIMARY | t_user | ALL | NULL | NULL | NULL | NULL | 1 | Using where |
| 2 | DEPENDENT SUBQUERY | t_user | ALL | NULL | NULL | NULL | NULL | 1 | Using where |
| 3 | DEPENDENT UNION | t_user | ALL | NULL | NULL | NULL | NULL | 1 | Using where |
| NULL | UNION RESULT | <union2,3> | ALL | NULL | NULL | NULL | NULL | NULL | Using temporary |
+----+--------------------+------------+------+---------------+------+---------+------+------+-----------------+
union result
包含union的结果集，在union和union all语句中,因为它不需要参与查询，所以id字段为null
derived
from字句中出现的子查询，也叫做派生表，其他数据库中可能叫做内联视图或嵌套select
mysql> explain select * from (select * from t_user where sex='1') b;
+----+-------------+------------+------+---------------+------+---------+------+------+-------------+
| id | select_type | table | type | possible_keys | key | key_len | ref | rows | Extra |
+----+-------------+------------+------+---------------+------+---------+------+------+-------------+
| 1 | PRIMARY | <derived2> | ALL | NULL | NULL | NULL | NULL | 2 | NULL |
| 2 | DERIVED | t_user | ALL | NULL | NULL | NULL | NULL | 1 | Using where |
+----+-------------+------------+------+---------------+------+---------+------+------+-------------+
table
- 显示的查询表名，如果查询使用了别名，那么这里显示的是别名
- 如果不涉及对数据表的操作，那么这显示为null
- 如果显示为尖括号括起来的就表示这个是临时表，后边的N就是执行计划中的id，表示结果来自于 这个查询产生。
如果是尖括号括起来的<union M,N>，与类似，也是一个临时表，表示这个结果来自于union查 询的id为M,N的结果集。
type总结
依次从好到差:
system，const，eq_ref，ref，fulltext，ref_or_null，unique_subquery， index_subquery，range，index_merge，index，ALL
除了all之外，其他的type都可以使用到索引，除了index_merge之外，其他的type只可以用到一个索 引。
优化器会选用最优索引 一个
注意事项:
最少要索引使用到range级别。
Type列详解
system
表只有一行记录（等于系统表），这是const类型的特列，平时不大会出现，可以忽略
mysql> explain select * from (select * from t_user where id=1) a;
+----+-------------+------------+--------+---------------+------+---------+------+------+--------------------------------+
| id | select_type | table | type | possible_keys | key | key_len | ref | rows | Extra |
+----+-------------+------------+--------+---------------+------+---------+------+------+--------------------------------+
| 1 | PRIMARY | <derived2> | system | NULL | NULL | NULL | NULL | 0 | const row not found |
| 2 | DERIVED | NULL | NULL | NULL | NULL | NULL | NULL | NULL | no matching row in const table |
+----+-------------+------------+--------+---------------+------+---------+------+------+--------------------------------+
const(重要)
使用唯一索引或者主键，返回记录一定是1行记录的等值where条件时，通常type是const。其他数据库
也叫做唯一索引扫描
mysql> explain select * from t_user where id=1;
+----+-------------+-------+------+---------------+------+---------+------+------+-----------------------------------------------------+
| id | select_type | table | type | possible_keys | key | key_len | ref | rows | Extra |
+----+-------------+-------+------+---------------+------+---------+------+------+-----------------------------------------------------+
| 1 | SIMPLE | NULL | NULL | NULL | NULL | NULL | NULL | NULL | Impossible WHERE noticed after reading const tables |
+----+-------------+-------+------+---------------+------+---------+------+------+-----------------------------------------------------+
eq_ref(重要)
关键字:连接字段主键或者唯一性索引。
此类型通常出现在多表的 join 查询, 表示对于前表的每一个结果, 都只能匹配到后表的一行结果. 并且查 询的比较操作通常是 ‘=’, 查询效率较高.
mysql> explain select a.id from t_user a left join t_department b on a.department_id=b.id;
+----+-------------+-------+--------+---------------+---------+---------+------------------------+------+-------------+
| id | select_type | table | type | possible_keys | key | key_len | ref | rows | Extra |
+----+-------------+-------+--------+---------------+---------+---------+------------------------+------+-------------+
| 1 | SIMPLE | a | index | NULL | idx_dep | 5 | NULL | 1 | Using index |
| 1 | SIMPLE | b | eq_ref | PRIMARY | PRIMARY | 4 | sample.a.department_id | 1 | Using index |
+----+-------------+-------+--------+---------------+---------+---------+------------------------+------+-------------+
select * from a,b where a.id=b.id (等值连接)
select * from a where name=‘zs’ (条件查询)
ref(重要)
针对非唯一性索引，使用等值(=)查询非主键。或者是使用了最左前缀规则索引的查询。
-- 非唯一索引
mysql> explain select * from t_user where department_id=1;
+----+-------------+--------+------+---------------+---------+---------+-------+------+-------+
| id | select_type | table | type | possible_keys | key | key_len | ref | rows | Extra |
+----+-------------+--------+------+---------------+---------+---------+-------+------+-------+
| 1 | SIMPLE | t_user | ref | idx_dep | idx_dep | 5 | const | 1 | NULL |
+----+-------------+--------+------+---------------+---------+---------+-------+------+-------+
-- 等值非主键连接
mysql> explain select a.id from t_user a left join t_department b on a.name=b.name;
+----+-------------+-------+------+---------------+------+---------+------+------+----------------------------------------------------+
| id | select_type | table | type | possible_keys | key | key_len | ref | rows | Extra |
+----+-------------+-------+------+---------------+------+---------+------+------+----------------------------------------------------+
| 1 | SIMPLE | a | ALL | NULL | NULL | NULL | NULL | 1 | NULL |
| 1 | SIMPLE | b | ALL | NULL | NULL | NULL | NULL | 1 | Using where; Using join buffer (Block Nested Loop) |
+----+-------------+-------+------+---------------+------+---------+------+------+----------------------------------------------------+
-- 最左前缀
mysql> explain select * from t_user where name = 'kojon';
+----+-------------+--------+------+---------------+------+---------+------+------+-------------+
| id | select_type | table | type | possible_keys | key | key_len | ref | rows | Extra |
+----+-------------+--------+------+---------------+------+---------+------+------+-------------+
| 1 | SIMPLE | t_user | ALL | NULL | NULL | NULL | NULL | 1 | Using where |
+----+-------------+--------+------+---------------+------+---------+------+------+-------------+
fulltext
全文索引检索，要注意，全文索引的优先级很高，若全文索引和普通索引同时存在时，mysql不管代价，优先选择使用全文索引
# fulltext
mysql> explain select * from t_address where match(addr) against('beijing');
+----+-------------+-----------+----------+---------------+------------+---------+------+------+-------------+
| id | select_type | table | type | possible_keys | key | key_len | ref | rows | Extra |
+----+-------------+-----------+----------+---------------+------------+---------+------+------+-------------+
| 1 | SIMPLE | t_address | fulltext | ft_address | ft_address | 0 | NULL | 1 | Using where |
+----+-------------+-----------+----------+---------------+------------+---------+------+------+-------------+
1 row in set (0.00 sec)
ref_or_null
与ref方法类似，只是增加了null值的比较。实际用的不多。
unique_subquery
用于where中的in形式子查询，子查询返回不重复值唯一值
index_subquery
用于in形式子查询使用到了辅助索引或者in常数列表，子查询可能返回重复值，可以使用索引将子查询 去重。
range(重要)
索引范围扫描，常见于使用>,<,is null,between ,in ,like等运算符的查询中。
mysql> # range(重要)
mysql> explain select * from t_user where id>1;
+----+-------------+--------+-------+---------------+---------+---------+------+------+-------------+
| id | select_type | table | type | possible_keys | key | key_len | ref | rows | Extra |
+----+-------------+--------+-------+---------------+---------+---------+------+------+-------------+
| 1 | SIMPLE | t_user | range | PRIMARY | PRIMARY | 4 | NULL | 1 | Using where |
+----+-------------+--------+-------+---------------+---------+---------+------+------+-------------+
1 row in set (0.01 sec)
mysql> -- like 前缀索引
mysql> explain select * from t_user where name like '%x';
+----+-------------+--------+------+---------------+------+---------+------+------+-------------+
| id | select_type | table | type | possible_keys | key | key_len | ref | rows | Extra |
+----+-------------+--------+------+---------------+------+---------+------+------+-------------+
| 1 | SIMPLE | t_user | ALL | NULL | NULL | NULL | NULL | 1 | Using where |
+----+-------------+--------+------+---------------+------+---------+------+------+-------------+
1 row in set (0.00 sec)
mysql> -- like 后缀索引
mysql> explain select * from t_user where login_name like 'x%';
+----+-------------+--------+-------+----------------+----------------+---------+------+------+-----------------------+
| id | select_type | table | type | possible_keys | key | key_len | ref | rows | Extra |
+----+-------------+--------+-------+----------------+----------------+---------+------+------+-----------------------+
| 1 | SIMPLE | t_user | range | idx_login_name | idx_login_name | 403 | NULL | 1 | Using index condition |
+----+-------------+--------+-------+----------------+----------------+---------+------+------+-----------------------+
1 row in set (0.01 sec)
index_merge
表示查询使用了两个以上的索引，最后取交集或者并集，常见and ，or的条件使用了不同的索引，官方 排序这个在ref_or_null之后，但是实际上由于要读取所个索引，性能可能大部分时间都不如range
index(重要)
关键字:条件是出现在索引树中的节点的。可能没有完全匹配索引。
索引全表扫描，把索引从头到尾扫一遍，常见于使用索引列就可以处理不需要读取数据文件的查询、可 以使用索引排序或者分组的查询。
mysql> -- 单索引
mysql> explain select login_name from t_user;
+----+-------------+--------+-------+---------------+----------------+---------+------+------+-------------+
| id | select_type | table | type | possible_keys | key | key_len | ref | rows | Extra |
+----+-------------+--------+-------+---------------+----------------+---------+------+------+-------------+
| 1 | SIMPLE | t_user | index | NULL | idx_login_name | 403 | NULL | 1 | Using index |
+----+-------------+--------+-------+---------------+----------------+---------+------+------+-------------+
mysql> -- 组合索引
mysql> explain select age from t_user;
+----+-------------+--------+------+---------------+------+---------+------+------+-------+
| id | select_type | table | type | possible_keys | key | key_len | ref | rows | Extra |
+----+-------------+--------+------+---------------+------+---------+------+------+-------+
| 1 | SIMPLE | t_user | ALL | NULL | NULL | NULL | NULL | 1 | NULL |
+----+-------------+--------+------+---------------+------+---------+------+------+-------+
mysql> #需要通过覆盖索引将ALL类型变成index
mysql> explain select login_name,age from t_user;
+----+-------------+--------+------+---------------+------+---------+------+------+-------+
| id | select_type | table | type | possible_keys | key | key_len | ref | rows | Extra |
+----+-------------+--------+------+---------------+------+---------+------+------+-------+
| 1 | SIMPLE | t_user | ALL | NULL | NULL | NULL | NULL | 1 | NULL |
+----+-------------+--------+------+---------------+------+---------+------+------+-------+
all(重要)
这个就是全表扫描数据文件，然后再在server层进行过滤返回符合要求的记录。
mysql> # all(重要)
mysql> explain select * from t_user;
+----+-------------+--------+------+---------------+------+---------+------+------+-------+
| id | select_type | table | type | possible_keys | key | key_len | ref | rows | Extra |
+----+-------------+--------+------+---------------+------+---------+------+------+-------+
| 1 | SIMPLE | t_user | ALL | NULL | NULL | NULL | NULL | 1 | NULL |
+----+-------------+--------+------+---------------+------+---------+------+------+-------+
回表查询
思考:如何使用索引?
possible_keys
此次查询中可能选用的索引，一个或多个
key
查询真正使用到的索引，select_type为index_merge时，这里可能出现两个以上的索引，其他的 select_type这里只会出现一个。
key_len
- 用于处理查询的索引长度，如果是单列索引，那就整个索引长度算进去，如果是多列索引，那么查
询不一定都能使用到所有的列，具体使用到了多少个列的索引，这里就会计算进去，没有使用到的
列，这里不会计算进去。 - 留意下这个列的值，算一下你的多列索引总长度就知道有没有使用到所有的列了。
- 另外，key_len只计算where条件用到的索引长度，而排序和分组就算用到了索引，也不会计算到key_len中。
看组合索引的使用情况
ref
如果是使用的常数等值查询，这里会显示const
如果是连接查询，被驱动表的执行计划这里会显示驱动表的关联字段
如果是条件使用了表达式或者函数，或者条件列发生了内部隐式转换，这里可能显示为func
rows
这里是执行计划中估算的扫描行数，不是精确值（InnoDB不是精确的值，MyISAM是精确的值，主要原因是InnoDB里面使用了MVCC并发机制）
extra（重要）
这个列包含不适合在其他列中显示单十分重要的额外的信息，这个列可以显示的信息非常多，有几十种，常用的有
no tables used
不带from字句的查询或者From dual查询
使用not in()形式子查询或not exists运算符的连接查询，这种叫做反连接
即，一般连接查询是先查询内表，再查询外表，反连接就是先查询外表，再查询内表。
using filesort（重要）
排序时无法使用到索引时，就会出现这个。常见于order by和group by语句中说明MySQL会使用一个外部的索引排序，而不是按照索引顺序进行读取。 MySQL中无法利用索引完成的排序操作称为“文件排序”
mysql> # using filesort(重要)
mysql> explain select * from t_user order by name;
+----+-------------+--------+------+---------------+------+---------+------+------+----------------+
| id | select_type | table | type | possible_keys | key | key_len | ref | rows | Extra |
+----+-------------+--------+------+---------------+------+---------+------+------+----------------+
| 1 | SIMPLE | t_user | ALL | NULL | NULL | NULL | NULL | 1 | Using filesort |
+----+-------------+--------+------+---------------+------+---------+------+------+----------------+
using index（重要）
查询时不需要回表查询，直接通过索引就可以获取查询的数据。
- 表示相应的SELECT查询中使用到了覆盖索引（Covering Index），避免访问表的数据行，效率不错！
- 如果同时出现Using Where ，说明索引被用来执行查找索引键值
- 如果没有同时出现Using Where ，表明索引用来读取数据而非执行查找动作。
mysql> # 全值匹配 覆盖索引
mysql> explain select name,age,sex from t_user ;
+----+-------------+--------+------+---------------+------+---------+------+------+-------+
| id | select_type | table | type | possible_keys | key | key_len | ref | rows | Extra |
+----+-------------+--------+------+---------------+------+---------+------+------+-------+
| 1 | SIMPLE | t_user | ALL | NULL | NULL | NULL | NULL | 1 | NULL |
+----+-------------+--------+------+---------------+------+---------+------+------+-------+
using temporary
表示使用了临时表存储中间结果。
MySQL在对查询结果order by和group by时使用临时表
临时表可以是内存临时表和磁盘临时表，执行计划中看不出来，需要查看status变量， used_tmp_table，used_tmp_disk_table才能看出来。
distinct
在select部分使用了distinct关键字 （索引字段）
mysql> explain select distinct a.id from t_user a,t_department b where a.department_id=b.id;
+----+-------------+-------+--------+--------------------------------+---------+---------+------------------------+------+-------------------------------------------+
| id | select_type | table | type | possible_keys | key | key_len | ref | rows | Extra |
+----+-------------+-------+--------+--------------------------------+---------+---------+------------------------+------+-------------------------------------------+
| 1 | SIMPLE | a | index | PRIMARY,idx_login_name,idx_dep | idx_dep | 5 | NULL | 1 | Using where; Using index; Using temporary |
| 1 | SIMPLE | b | eq_ref | PRIMARY | PRIMARY | 4 | sample.a.department_id | 1 | Using index; Distinct |
+----+-------------+-------+--------+--------------------------------+---------+---------+------------------------+------+-------------------------------------------+
using where（重要）
表示存储引擎返回的记录并不是所有的都满足查询条件，需要在server层进行过滤。
mysql> -- 查询条件无索引
mysql> explain select * from t_user where address='beijing';
+----+-------------+--------+------+---------------+------+---------+------+------+-------------+
| id | select_type | table | type | possible_keys | key | key_len | ref | rows | Extra |
+----+-------------+--------+------+---------------+------+---------+------+------+-------------+
| 1 | SIMPLE | t_user | ALL | NULL | NULL | NULL | NULL | 1 | Using where |
+----+-------------+--------+------+---------------+------+---------+------+------+-------------+
1 row in set (0.00 sec)
mysql> -- 索引失效
mysql> explain select * from t_user where age=1;
+----+-------------+--------+------+---------------+------+---------+------+------+-------------+
| id | select_type | table | type | possible_keys | key | key_len | ref | rows | Extra |
+----+-------------+--------+------+---------------+------+---------+------+------+-------------+
| 1 | SIMPLE | t_user | ALL | NULL | NULL | NULL | NULL | 1 | Using where |
+----+-------------+--------+------+---------------+------+---------+------+------+-------------+
1 row in set (0.00 sec)
mysql> -- 索引失效
mysql> explain select * from t_user where id in(1,2);
+----+-------------+--------+-------+---------------+---------+---------+------+------+-------------+
| id | select_type | table | type | possible_keys | key | key_len | ref | rows | Extra |
+----+-------------+--------+-------+---------------+---------+---------+------+------+-------------+
| 1 | SIMPLE | t_user | range | PRIMARY | PRIMARY | 4 | NULL | 2 | Using where |
+----+-------------+--------+-------+---------------+---------+---------+------+------+-------------+
1 row in set (0.00 sec)
查询条件中分为限制条件和检查条件，5.6之前，存储引擎只能根据限制条件扫描数据并返回，然后server层根据检查条件进行过滤再返回真正符合查询的数据。5.6.x之后支持ICP特性，可以把检查条件也下推到存储引擎层，不符合检查条件和限制条件的数据，直接不读取，这样就大大减少了存储引擎扫描的记录数量。extra列显示using index condition
mysql> -- 索引下推
mysql> explain select * from t_user where name='abc';
+----+-------------+--------+------+---------------+------+---------+------+------+-------------+
| id | select_type | table | type | possible_keys | key | key_len | ref | rows | Extra |
+----+-------------+--------+------+---------------+------+---------+------+------+-------------+
| 1 | SIMPLE | t_user | ALL | NULL | NULL | NULL | NULL | 1 | Using where |
+----+-------------+--------+------+---------------+------+---------+------+------+-------------+
索引失效分析
1.全值匹配我最爱
mysql> explain select * from t_user where name='kojon' and age=1 and sex='1';
2.最佳左前缀法组合索引
组合索引
带头索引不能死，中间索引不能断
如果索引了多个列，要遵守最佳左前缀法则。指的是查询从索引的最左前列开始 并且不跳过索引中的列。
错误的示例：
带头索引死：
mysql> -- 带头索引死
mysql> explain select * from t_user where age=23;
+----+-------------+--------+------+---------------+------+---------+------+------+-------------+
| id | select_type | table | type | possible_keys | key | key_len | ref | rows | Extra |
+----+-------------+--------+------+---------------+------+---------+------+------+-------------+
| 1 | SIMPLE | t_user | ALL | NULL | NULL | NULL | NULL | 1 | Using where |
+----+-------------+--------+------+---------------+------+---------+------+------+-------------+
1 row in set (0.00 sec)
中间索引断（带头索引生效，其他索引失效）：
mysql> -- 中间索引断
mysql> explain select * from t_user where name='aa' and sex='1';
+----+-------------+--------+------+---------------+------+---------+------+------+-------------+
| id | select_type | table | type | possible_keys | key | key_len | ref | rows | Extra |
+----+-------------+--------+------+---------------+------+---------+------+------+-------------+
| 1 | SIMPLE | t_user | ALL | NULL | NULL | NULL | NULL | 1 | Using where |
+----+-------------+--------+------+---------------+------+---------+------+------+-------------+
比较
mysql> explain select * from t_user where name='aa' and sex='1' and age=23;
+----+-------------+--------+------+---------------+------+---------+------+------+-------------+
| id | select_type | table | type | possible_keys | key | key_len | ref | rows | Extra |
+----+-------------+--------+------+---------------+------+---------+------+------+-------------+
| 1 | SIMPLE | t_user | ALL | NULL | NULL | NULL | NULL | 1 | Using where |
+----+-------------+--------+------+---------------+------+---------+------+------+-------------+
mysql> explain select * from t_user where name='aa' and sex=1 and age=23;
+----+-------------+--------+------+---------------+------+---------+------+------+-------------+
| id | select_type | table | type | possible_keys | key | key_len | ref | rows | Extra |
+----+-------------+--------+------+---------------+------+---------+------+------+-------------+
| 1 | SIMPLE | t_user | ALL | NULL | NULL | NULL | NULL | 1 | Using where |
+----+-------------+--------+------+---------------+------+---------+------+------+-------------+
3.不要在索引上做计算
mysql> #不要进行这些操作:计算、函数、自动/手动类型转换，不然会导致 索引失效而转向全表扫描
mysql> explain select * from t_user where login_name='xh';
+----+-------------+-------+------+---------------+------+---------+------+------+-----------------------------------------------------+
| id | select_type | table | type | possible_keys | key | key_len | ref | rows | Extra |
+----+-------------+-------+------+---------------+------+---------+------+------+-----------------------------------------------------+
| 1 | SIMPLE | NULL | NULL | NULL | NULL | NULL | NULL | NULL | Impossible WHERE noticed after reading const tables |
+----+-------------+-------+------+---------------+------+---------+------+------+-----------------------------------------------------+
1 row in set (0.00 sec)
mysql> -- 不要在索引上做计算
mysql> explain select * from t_user where left(login_name,1)='xh';
+----+-------------+--------+------+---------------+------+---------+------+------+-------------+
| id | select_type | table | type | possible_keys | key | key_len | ref | rows | Extra |
+----+-------------+--------+------+---------------+------+---------+------+------+-------------+
| 1 | SIMPLE | t_user | ALL | NULL | NULL | NULL | NULL | 1 | Using where |
+----+-------------+--------+------+---------------+------+---------+------+------+-------------+
4.范围条件右边的列失效
mysql> -- 范围条件右边的列失效
mysql> explain select * from t_user where name='abc' and age>20 and sex='1';
+----+-------------+--------+------+---------------+------+---------+------+------+-------------+
| id | select_type | table | type | possible_keys | key | key_len | ref | rows | Extra |
+----+-------------+--------+------+---------------+------+---------+------+------+-------------+
| 1 | SIMPLE | t_user | ALL | NULL | NULL | NULL | NULL | 1 | Using where |
+----+-------------+--------+------+---------------+------+---------+------+------+-------------+
5.尽量使用覆盖索引
mysql> -- 尽量使用覆盖索引(只查询索引的列)，也就是索引列和查询列一 致，减少select *
mysql> explain select * from t_user ;
+----+-------------+--------+------+---------------+------+---------+------+------+-------+
| id | select_type | table | type | possible_keys | key | key_len | ref | rows | Extra |
+----+-------------+--------+------+---------------+------+---------+------+------+-------+
| 1 | SIMPLE | t_user | ALL | NULL | NULL | NULL | NULL | 1 | NULL |
+----+-------------+--------+------+---------------+------+---------+------+------+-------+
mysql> -- 未使用覆盖索引
mysql> explain select name,login_name from t_user ;
+----+-------------+--------+------+---------------+------+---------+------+------+-------+
| id | select_type | table | type | possible_keys | key | key_len | ref | rows | Extra |
+----+-------------+--------+------+---------------+------+---------+------+------+-------+
| 1 | SIMPLE | t_user | ALL | NULL | NULL | NULL | NULL | 1 | NULL |
+----+-------------+--------+------+---------------+------+---------+------+------+-------+
1 row in set (0.00 sec)
mysql> -- 使用覆盖索引
mysql> explain select name,age,sex from t_user ;
+----+-------------+--------+------+---------------+------+---------+------+------+-------+
| id | select_type | table | type | possible_keys | key | key_len | ref | rows | Extra |
+----+-------------+--------+------+---------------+------+---------+------+------+-------+
| 1 | SIMPLE | t_user | ALL | NULL | NULL | NULL | NULL | 1 | NULL |
+----+-------------+--------+------+---------------+------+---------+------+------+-------+
1 row in set (0.00 sec)
mysql> -- 使用索引
mysql> explain select login_name from t_user ;
+----+-------------+--------+-------+---------------+----------------+---------+------+------+-------------+
| id | select_type | table | type | possible_keys | key | key_len | ref | rows | Extra |
+----+-------------+--------+-------+---------------+----------------+---------+------+------+-------------+
| 1 | SIMPLE | t_user | index | NULL | idx_login_name | 403 | NULL | 1 | Using index |
+----+-------------+--------+-------+---------------+----------------+---------+------+------+-------------+
6.索引字段上不要使用不等
mysql> -- 索引字段上使用等于
mysql> explain select * from t_user where login_name='xh';
+----+-------------+-------+------+---------------+------+---------+------+------+-----------------------------------------------------+
| id | select_type | table | type | possible_keys | key | key_len | ref | rows | Extra |
+----+-------------+-------+------+---------------+------+---------+------+------+-----------------------------------------------------+
| 1 | SIMPLE | NULL | NULL | NULL | NULL | NULL | NULL | NULL | Impossible WHERE noticed after reading const tables |
+----+-------------+-------+------+---------------+------+---------+------+------+-----------------------------------------------------+
1 row in set (0.00 sec)
mysql> -- 索引字段上使用不等于
mysql> explain select * from t_user where login_name!='xh';
+----+-------------+--------+------+----------------+------+---------+------+------+-------------+
| id | select_type | table | type | possible_keys | key | key_len | ref | rows | Extra |
+----+-------------+--------+------+----------------+------+---------+------+------+-------------+
| 1 | SIMPLE | t_user | ALL | idx_login_name | NULL | NULL | NULL | 1 | Using where |
+----+-------------+--------+------+----------------+------+---------+------+------+-------------+
7.主键索引字段上不可以判断null
mysql> # 主键字段上不可以使用 null
mysql> explain select * from t_user where name is null;
+----+-------------+--------+------+---------------+------+---------+------+------+-------------+
| id | select_type | table | type | possible_keys | key | key_len | ref | rows | Extra |
+----+-------------+--------+------+---------------+------+---------+------+------+-------------+
| 1 | SIMPLE | t_user | ALL | NULL | NULL | NULL | NULL | 1 | Using where |
+----+-------------+--------+------+---------------+------+---------+------+------+-------------+
1 row in set (0.00 sec)
mysql> # 索引字段上使用 is null 判断时，可使用索引
mysql> explain select * from t_user where login_name is null;
+----+-------------+--------+------+----------------+----------------+---------+-------+------+-----------------------+
| id | select_type | table | type | possible_keys | key | key_len | ref | rows | Extra |
+----+-------------+--------+------+----------------+----------------+---------+-------+------+-----------------------+
| 1 | SIMPLE | t_user | ref | idx_login_name | idx_login_name | 403 | const | 1 | Using index condition |
+----+-------------+--------+------+----------------+----------------+---------+-------+------+-----------------------+
mysql> -- 主键非空 不使用索引
mysql> explain select * from t_user where id is not null;
+----+-------------+--------+------+---------------+------+---------+------+------+-------------+
| id | select_type | table | type | possible_keys | key | key_len | ref | rows | Extra |
+----+-------------+--------+------+---------------+------+---------+------+------+-------------+
| 1 | SIMPLE | t_user | ALL | PRIMARY | NULL | NULL | NULL | 1 | Using where |
+----+-------------+--------+------+---------------+------+---------+------+------+-------------+
8.索引字段使用like不以通配符开头
mysql> -- 不以%开头
mysql> explain select * from t_user where login_name like 'x%';
+----+-------------+--------+-------+----------------+----------------+---------+------+------+-----------------------+
| id | select_type | table | type | possible_keys | key | key_len | ref | rows | Extra |
+----+-------------+--------+-------+----------------+----------------+---------+------+------+-----------------------+
| 1 | SIMPLE | t_user | range | idx_login_name | idx_login_name | 403 | NULL | 1 | Using index condition |
+----+-------------+--------+-------+----------------+----------------+---------+------+------+-----------------------+
1 row in set (0.00 sec)
mysql> -- 索引字段使用like以通配符开头(‘%字符串’)时，会导致索引失效而转向全表扫描
mysql> explain select * from t_user where login_name like '%x';
+----+-------------+--------+------+---------------+------+---------+------+------+-------------+
| id | select_type | table | type | possible_keys | key | key_len | ref | rows | Extra |
+----+-------------+--------+------+---------------+------+---------+------+------+-------------+
| 1 | SIMPLE | t_user | ALL | NULL | NULL | NULL | NULL | 1 | Using where |
+----+-------------+--------+------+---------------+------+---------+------+------+-------------+
1 row in set (0.00 sec)
由结果可知，like以通配符结束相当于范围查找，索引不会失效。与范围条件（bettween、<、>、in等）不同的是：不会导致右边的索引失效。
9.索引字段字符串要加单引号
隐式转换问题
索引字段是字符串，但查询时不加单引号，会导致索引失效而转向全表扫描
mysql> # 索引字段字符串要加单引号
mysql> explain select * from t_user where name=123;
+----+-------------+--------+------+---------------+------+---------+------+------+-------------+
| id | select_type | table | type | possible_keys | key | key_len | ref | rows | Extra |
+----+-------------+--------+------+---------------+------+---------+------+------+-------------+
| 1 | SIMPLE | t_user | ALL | NULL | NULL | NULL | NULL | 1 | Using where |
+----+-------------+--------+------+---------------+------+---------+------+------+-------------+
10.索引字段不要使用or
索引字段使用 or 时，会导致索引失效而转向全表扫描
mysql> # 索引字段不要使用or
mysql> explain select * from t_user where name='asd' or age=18;
+----+-------------+--------+------+---------------+------+---------+------+------+-------------+
| id | select_type | table | type | possible_keys | key | key_len | ref | rows | Extra |
+----+-------------+--------+------+---------------+------+---------+------+------+-------------+
| 1 | SIMPLE | t_user | ALL | NULL | NULL | NULL | NULL | 1 | Using where |
+----+-------------+--------+------+---------------+------+---------+------+------+-------------+
总结
[优化总结口诀]
全值匹配我最爱，最左前綴要遵守;
带头大哥不能死，中间兄弟不能断;
索引列上少计算，范围之后全失效;
LIKE百分写最右，覆盖索引不写星;
不等空值还有or，索引失效要少用.
扫一扫
被折叠的 条评论
为什么被折叠?
到【灌水乐园】发言
