MySql索引详解

1、索引

索引是表的目录，在查找内容之前可以先在目录中查找索引位置，以此快速定位查询数据。对于索引，会保存在额外的文件中。

2、索引，是数据库中专门用于帮助用户快速查询数据的一种数据结构。类似于字典中的目录，查找字典内容时可以根据目录查找到数据的存放位置，然后直接获取即可。

索引由数据库中一列或多列组合而成，其作用是提高对表中数据的查询速度
索引的优点是可以提高检索数据的速度
索引的缺点是创建和维护索引需要耗费时间
索引可以提高查询速度，会减慢写入速度

索引分类

1.普通索引
2.唯一索引
3.全文索引
4.单列索引
5.多列索引
6.空间索引
7.主键索引
8.组合索引

• 普通索引：仅加速查询
• 唯一索引：加速查询 + 列值唯一（可以有null）
• 主键索引：加速查询 + 列值唯一 +　表中只有一个（不可以有null）
• 组合索引：多列值组成一个索引，
  专门用于组合搜索，其效率大于索引合并
• 全文索引：对文本的内容进行分词，进行搜索

索引合并，使用多个单列索引组合搜索
覆盖索引，select的数据列只用从索引中就能够取得，不必读取数据行，换句话说查询列要被所建的索引覆盖

如何创建索引？记住一个单词—explain

创建表的时候创建索引

1 2 3 4 5 6

<span class="hljs-keyword">CREATE <span class="hljs-keyword">TABLE tbl_name( 字段名称 字段类型 [完整性约束条件]， ，，，, [<span class="hljs-keyword">UNIQUE|FULLTEXT|SPATIAL] <span class="hljs-keyword">INDEX|<span class="hljs-keyword">KEY[索引名称](字段名称[(长度)]) [<span class="hljs-keyword">ASC|<span class="hljs-keyword">DESC] );</span></span></span></span></span></span></span>

在已经存在的表上创建索引：

1 2	1.CREATE [UNIQUE|FULLTEXT|SPATIAL] INDEX 索引名称 ON 表名{字段名称[(长度)] [ASC|DESC]} 2.ALTER TABLE tbl_name ADD [<span class="hljs-string">UNIQUE|FULLTEXT|SPATIAL] INDEX索引名称(字段名称[<span class="hljs-string">(长度)][<span class="hljs-symbol">ASC|DESC]);</span></span></span>

如何删除索引？

1	<span class="hljs-keyword">DROP <span class="hljs-keyword">INDEX 索引名称 <span class="hljs-keyword">ON tbl_name</span></span></span>

1.普通索引

普通索引仅有一个功能：加速查询

创建表+索引

1 2 3 4 5 6 7

<span class="hljs-keyword">create <span class="hljs-keyword">table in1(     nid <span class="hljs-built_in">int <span class="hljs-keyword">not <span class="hljs-literal">null auto_increment primary <span class="hljs-keyword">key,     <span class="hljs-keyword">name <span class="hljs-built_in">varchar(<span class="hljs-number">32) <span class="hljs-keyword">not <span class="hljs-literal">null,     email <span class="hljs-built_in">varchar(<span class="hljs-number">64) <span class="hljs-keyword">not <span class="hljs-literal">null,     extra <span class="hljs-built_in">text,     <span class="hljs-keyword">index ix_name (<span class="hljs-keyword">name) )</span></span></span></span></span></span></span></span></span></span></span></span></span></span></span></span></span></span>

创建索引

1	<span class="hljs-keyword">create <span class="hljs-keyword">index index_name <span class="hljs-keyword">on table_name(column_name)</span></span></span>

删除索引

1	<span class="hljs-keyword">drop index_name <span class="hljs-keyword">on table_name;</span></span>

查看索引

1	<span class="hljs-keyword">show <span class="hljs-keyword">index <span class="hljs-keyword">from table_name;</span></span></span>

注意：对于创建索引时如果是BLOB 和 TEXT 类型，必须指定length。

1	<span class="hljs-keyword">create <span class="hljs-keyword">index ix_extra <span class="hljs-keyword">on in1(extra(<span class="hljs-number">32));</span></span></span></span>

2、唯一索引

唯一索引有两个功能：加速查询 和 唯一约束（可含null）

创建表+唯一索引

1 2 3 4 5 6 7

<span class="hljs-keyword">create <span class="hljs-keyword">table in1(     nid <span class="hljs-built_in">int <span class="hljs-keyword">not <span class="hljs-literal">null auto_increment primary <span class="hljs-keyword">key,     <span class="hljs-keyword">name <span class="hljs-built_in">varchar(<span class="hljs-number">32) <span class="hljs-keyword">not <span class="hljs-literal">null,     email <span class="hljs-built_in">varchar(<span class="hljs-number">64) <span class="hljs-keyword">not <span class="hljs-literal">null,     extra <span class="hljs-built_in">text,     <span class="hljs-keyword">unique ix_name (<span class="hljs-keyword">name) )</span></span></span></span></span></span></span></span></span></span></span></span></span></span></span></span></span></span>

创建唯一索引

1	<span class="hljs-keyword">create <span class="hljs-keyword">unique <span class="hljs-keyword">index 索引名 <span class="hljs-keyword">on 表名(列名)</span></span></span></span>

删除唯一索引

1	<span class="hljs-keyword">drop <span class="hljs-keyword">unique <span class="hljs-keyword">index 索引名 <span class="hljs-keyword">on 表名</span></span></span></span>

3、主键索引

主键有两个功能：加速查询 和 唯一约束（不可含null）

创建表+创建主键

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18

<span class="hljs-keyword">create <span class="hljs-keyword">table in1(     nid <span class="hljs-built_in">int <span class="hljs-keyword">not <span class="hljs-literal">null auto_increment primary <span class="hljs-keyword">key,     <span class="hljs-keyword">name <span class="hljs-built_in">varchar(<span class="hljs-number">32) <span class="hljs-keyword">not <span class="hljs-literal">null,     email <span class="hljs-built_in">varchar(<span class="hljs-number">64) <span class="hljs-keyword">not <span class="hljs-literal">null,     extra <span class="hljs-built_in">text,     <span class="hljs-keyword">index ix_name (<span class="hljs-keyword">name) )   <span class="hljs-keyword">OR   <span class="hljs-keyword">create <span class="hljs-keyword">table in1(     nid <span class="hljs-built_in">int <span class="hljs-keyword">not <span class="hljs-literal">null auto_increment,     <span class="hljs-keyword">name <span class="hljs-built_in">varchar(<span class="hljs-number">32) <span class="hljs-keyword">not <span class="hljs-literal">null,     email <span class="hljs-built_in">varchar(<span class="hljs-number">64) <span class="hljs-keyword">not <span class="hljs-literal">null,     extra <span class="hljs-built_in">text,     primary <span class="hljs-keyword">key(ni1),     <span class="hljs-keyword">index ix_name (<span class="hljs-keyword">name) )</span></span></span></span></span></span></span></span></span></span></span></span></span></span></span></span></span></span></span></span></span></span></span></span></span></span></span></span></span></span></span></span></span></span></span></span></span>

创建主键

1	<span class="hljs-keyword">alter <span class="hljs-keyword">table 表名 <span class="hljs-keyword">add primary <span class="hljs-keyword">key(列名);</span></span></span></span>

删除主键

1 2

<span class="hljs-keyword">alter <span class="hljs-keyword">table 表名 <span class="hljs-keyword">drop primary <span class="hljs-keyword">key; <span class="hljs-keyword">alter <span class="hljs-keyword">table 表名  <span class="hljs-keyword">modify  列名 <span class="hljs-built_in">int, <span class="hljs-keyword">drop primary <span class="hljs-keyword">key;</span></span></span></span></span></span></span></span></span></span>

4、组合索引

组合索引是将n个列组合成一个索引

其应用场景为：频繁的同时使用n列来进行查询，如：where n1 = ‘alex’ and n2 = 666。

创建表

1 2 3 4 5 6

<span class="hljs-keyword">create <span class="hljs-keyword">table in3(     nid <span class="hljs-built_in">int <span class="hljs-keyword">not <span class="hljs-literal">null auto_increment primary <span class="hljs-keyword">key,     <span class="hljs-keyword">name <span class="hljs-built_in">varchar(<span class="hljs-number">32) <span class="hljs-keyword">not <span class="hljs-literal">null,     email <span class="hljs-built_in">varchar(<span class="hljs-number">64) <span class="hljs-keyword">not <span class="hljs-literal">null,     extra <span class="hljs-built_in">text )</span></span></span></span></span></span></span></span></span></span></span></span></span></span></span></span>

创建组合索引

1	<span class="hljs-keyword">create <span class="hljs-keyword">index ix_name_email <span class="hljs-keyword">on in3(<span class="hljs-keyword">name,email);</span></span></span></span>

如上创建组合索引之后，查询：

• name and email  – 使用索引
• name                 — 使用索引
• email                 — 不使用索引

注意：对于同时搜索n个条件时，组合索引的性能好于多个单一索引合并。

相关命令

1 2 3 4 5 6 7 8 9 10 11 12 13

- 查看表结构     desc 表名   - 查看生成表的SQL     <span class="hljs-keyword">show <span class="hljs-keyword">create <span class="hljs-keyword">table 表名   - 查看索引     <span class="hljs-keyword">show <span class="hljs-keyword">index <span class="hljs-keyword">from  表名   - 查看执行时间     <span class="hljs-keyword">set profiling = <span class="hljs-number">1;     SQL...     <span class="hljs-keyword">show <span class="hljs-keyword">profiles;</span></span></span></span></span></span></span></span></span></span>

使用索引和不使用索引

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19

由于索引是专门用于加速搜索而生，所以加上索引之后，查询效率会快到飞起来。   <span class="hljs-comment"># 有索引 mysql> select * from tb1 where name = <span class="hljs-string">'wupeiqi-888'; +-----+-------------+---------------------+----------------------------------+---------------------+ <span class="hljs-params">| nid | name        <span class="hljs-params">| email               | radom                            <span class="hljs-params">| ctime               | +-----+-------------+---------------------+----------------------------------+---------------------+ <span class="hljs-params">| 889 | wupeiqi-<span class="hljs-number">888 <span class="hljs-params">| wupeiqi888@live.com | <span class="hljs-number">5312269e76a16a90b8a8301d5314204b <span class="hljs-params">| 2016-08-03 09:33:35 | +-----+-------------+---------------------+----------------------------------+---------------------+ <span class="hljs-number">1 row <span class="hljs-keyword">in set (<span class="hljs-number">0.<span class="hljs-number">00 sec)   <span class="hljs-comment"># 无索引 mysql> select * from tb1 where email = <span class="hljs-string">'wupeiqi888@live.com'; +-----+-------------+---------------------+----------------------------------+---------------------+ <span class="hljs-params">| nid | name        <span class="hljs-params">| email               | radom                            <span class="hljs-params">| ctime               | +-----+-------------+---------------------+----------------------------------+---------------------+ <span class="hljs-params">| 889 | wupeiqi-<span class="hljs-number">888 <span class="hljs-params">| wupeiqi888@live.com | <span class="hljs-number">5312269e76a16a90b8a8301d5314204b <span class="hljs-params">| 2016-08-03 09:33:35 | +-----+-------------+---------------------+----------------------------------+---------------------+ <span class="hljs-number">1 row <span class="hljs-keyword">in set (<span class="hljs-number">1.23 sec)</span></span></span></span></span></span></span></span></span></span></span></span></span></span></span></span></span></span></span></span></span></span></span></span></span></span></span>

正确使用索引

数据库表中添加索引后确实会让查询速度起飞，但前提必须是正确的使用索引来查询，如果以错误的方式使用，则即使建立索引也会不奏效。

即使建立索引，索引也不会生效：

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32

- like '%xx'     <span class="hljs-keyword">select * <span class="hljs-keyword">from tb1 <span class="hljs-keyword">where <span class="hljs-keyword">name <span class="hljs-keyword">like <span class="hljs-string">'%cn'; - 使用函数     <span class="hljs-keyword">select * <span class="hljs-keyword">from tb1 <span class="hljs-keyword">where <span class="hljs-keyword">reverse(<span class="hljs-keyword">name) = <span class="hljs-string">'wupeiqi'; - or     <span class="hljs-keyword">select * <span class="hljs-keyword">from tb1 <span class="hljs-keyword">where nid = <span class="hljs-number">1 <span class="hljs-keyword">or email = <span class="hljs-string">'seven@live.com';     特别的：当or条件中有未建立索引的列才失效，以下会走索引             <span class="hljs-keyword">select * <span class="hljs-keyword">from tb1 <span class="hljs-keyword">where nid = <span class="hljs-number">1 <span class="hljs-keyword">or <span class="hljs-keyword">name = <span class="hljs-string">'seven';             <span class="hljs-keyword">select * <span class="hljs-keyword">from tb1 <span class="hljs-keyword">where nid = <span class="hljs-number">1 <span class="hljs-keyword">or email = <span class="hljs-string">'seven@live.com' <span class="hljs-keyword">and <span class="hljs-keyword">name = <span class="hljs-string">'alex' - 类型不一致     如果列是字符串类型，传入条件是必须用引号引起来，不然...     <span class="hljs-keyword">select * <span class="hljs-keyword">from tb1 <span class="hljs-keyword">where <span class="hljs-keyword">name = <span class="hljs-number">999; - !=     <span class="hljs-keyword">select * <span class="hljs-keyword">from tb1 <span class="hljs-keyword">where <span class="hljs-keyword">name != <span class="hljs-string">'alex'     特别的：如果是主键，则还是会走索引         <span class="hljs-keyword">select * <span class="hljs-keyword">from tb1 <span class="hljs-keyword">wherenid != <span class="hljs-number">123 - >     <span class="hljs-keyword">select * <span class="hljs-keyword">from tb1 <span class="hljs-keyword">where <span class="hljs-keyword">name > <span class="hljs-string">'alex'     特别的：如果是主键或索引是整数类型，则还是会走索引         <span class="hljs-keyword">select * <span class="hljs-keyword">from tb1 <span class="hljs-keyword">wherenid > <span class="hljs-number">123         <span class="hljs-keyword">select * <span class="hljs-keyword">from tb1 <span class="hljs-keyword">where<span class="hljs-keyword">num > <span class="hljs-number">123 - <span class="hljs-keyword">order <span class="hljs-keyword">by     <span class="hljs-keyword">select email <span class="hljs-keyword">from tb1 <span class="hljs-keyword">order<span class="hljs-keyword">by <span class="hljs-keyword">name <span class="hljs-keyword">desc;     当根据索引排序时候，选择的映射如果不是索引，则不走索引     特别的：如果对主键排序，则还是走索引：         <span class="hljs-keyword">select * <span class="hljs-keyword">from tb1 <span class="hljs-keyword">order<span class="hljs-keyword">by nid <span class="hljs-keyword">desc;   - 组合索引最左前缀     如果组合索引为：(name,email)     name and email       <span class="hljs-comment">-- 使用索引     name                 <span class="hljs-comment">-- 使用索引     email                <span class="hljs-comment">-- 不使用索引</span></span></span></span></span></span></span></span></span></span></span></span></span></span></span></span></span></span></span></span></span></span></span></span></span></span></span></span></span></span></span></span></span></span></span></span></span></span></span></span></span></span></span></span></span></span></span></span></span></span></span></span></span></span></span></span></span></span></span></span></span></span></span></span></span></span></span></span></span></span></span></span></span></span></span></span></span></span>

其他注意事项

1 2 3 4 5 6 7 8 9

<span class="hljs-deletion">- 避免使用select * <span class="hljs-deletion">- count(1)或count(列) 代替 count(*) <span class="hljs-deletion">- 创建表时尽量时 char 代替 varchar <span class="hljs-deletion">- 表的字段顺序固定长度的字段优先 <span class="hljs-deletion">- 组合索引代替多个单列索引（经常使用多个条件查询时） <span class="hljs-deletion">- 尽量使用短索引 <span class="hljs-deletion">- 使用连接（JOIN）来代替子查询(Sub-Queries) <span class="hljs-deletion">- 连表时注意条件类型需一致 <span class="hljs-deletion">- 索引散列值（重复少）不适合建索引，例：性别不适合</span></span></span></span></span></span></span></span></span>

limit分页

无论是否有索引，limit分页是一个值得关注的问题

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51

每页显示10条： 当前 118 120， 125   倒序：             大      小             980    970  7 6  6 5  54  43  32   21 19 98     下一页：       <span class="hljs-keyword">select         *     <span class="hljs-keyword">from         tb1     <span class="hljs-keyword">where         nid < (<span class="hljs-keyword">select nid <span class="hljs-keyword">from (<span class="hljs-keyword">select nid <span class="hljs-keyword">from tb1 <span class="hljs-keyword">where nid < 当前页最小值 <span class="hljs-keyword">order <span class="hljs-keyword">by nid <span class="hljs-keyword">desc <span class="hljs-keyword">limit 每页数据 *【页码-当前页】) A <span class="hljs-keyword">order <span class="hljs-keyword">by A.nid <span class="hljs-keyword">asc <span class="hljs-keyword">limit <span class="hljs-number">1)      <span class="hljs-keyword">order <span class="hljs-keyword">by         nid <span class="hljs-keyword">desc     <span class="hljs-keyword">limit <span class="hljs-number">10;       <span class="hljs-keyword">select         *     <span class="hljs-keyword">from         tb1     <span class="hljs-keyword">where         nid < (<span class="hljs-keyword">select nid <span class="hljs-keyword">from (<span class="hljs-keyword">select nid <span class="hljs-keyword">from tb1 <span class="hljs-keyword">where nid < <span class="hljs-number">970  <span class="hljs-keyword">order <span class="hljs-keyword">by nid <span class="hljs-keyword">desc <span class="hljs-keyword">limit <span class="hljs-number">40) A <span class="hljs-keyword">order <span class="hljs-keyword">by A.nid <span class="hljs-keyword">asc <span class="hljs-keyword">limit <span class="hljs-number">1)      <span class="hljs-keyword">order <span class="hljs-keyword">by         nid <span class="hljs-keyword">desc     <span class="hljs-keyword">limit <span class="hljs-number">10;   上一页：       <span class="hljs-keyword">select         *     <span class="hljs-keyword">from         tb1     <span class="hljs-keyword">where         nid < (<span class="hljs-keyword">select nid <span class="hljs-keyword">from (<span class="hljs-keyword">select nid <span class="hljs-keyword">from tb1 <span class="hljs-keyword">where nid > 当前页最大值 <span class="hljs-keyword">order <span class="hljs-keyword">by nid <span class="hljs-keyword">asc <span class="hljs-keyword">limit 每页数据 *【当前页-页码】) A <span class="hljs-keyword">order <span class="hljs-keyword">by A.nid <span class="hljs-keyword">asc <span class="hljs-keyword">limit <span class="hljs-number">1)      <span class="hljs-keyword">order <span class="hljs-keyword">by         nid <span class="hljs-keyword">desc     <span class="hljs-keyword">limit <span class="hljs-number">10;       <span class="hljs-keyword">select         *     <span class="hljs-keyword">from         tb1     <span class="hljs-keyword">where         nid < (<span class="hljs-keyword">select nid <span class="hljs-keyword">from (<span class="hljs-keyword">select nid <span class="hljs-keyword">from tb1 <span class="hljs-keyword">where nid > <span class="hljs-number">980 <span class="hljs-keyword">order <span class="hljs-keyword">by nid <span class="hljs-keyword">asc <span class="hljs-keyword">limit <span class="hljs-number">20) A <span class="hljs-keyword">order <span class="hljs-keyword">by A.nid <span class="hljs-keyword">desc <span class="hljs-keyword">limit <span class="hljs-number">1)      <span class="hljs-keyword">order <span class="hljs-keyword">by         nid <span class="hljs-keyword">desc     <span class="hljs-keyword">limit <span class="hljs-number">10;</span></span></span></span></span></span></span></span></span></span></span></span></span></span></span></span></span></span></span></span></span></span></span></span></span></span></span></span></span></span></span></span></span></span></span></span></span></span></span></span></span></span></span></span></span></span></span></span></span></span></span></span></span></span></span></span></span></span></span></span></span></span></span></span></span></span></span></span></span></span></span></span></span></span></span></span></span></span></span></span></span></span></span></span></span></span></span></span></span></span></span></span>

执行计划

explain + 查询SQL – 用于显示SQL执行信息参数，根据参考信息可以进行SQL优化

1 2 3 4 5 6 7

mysql> explain select * from tb2; +----+-------------+-------+------+---------------+------+---------+------+------+-------+ <span class="hljs-params">| id | select_type <span class="hljs-params">| table | type <span class="hljs-params">| possible_keys | key  <span class="hljs-params">| key_len | ref  <span class="hljs-params">| rows | Extra <span class="hljs-params">| +----+-------------+-------+------+---------------+------+---------+------+------+-------+ |  <span class="hljs-number">1 <span class="hljs-params">| SIMPLE      | tb2   <span class="hljs-params">| ALL  | NULL          <span class="hljs-params">| NULL | NULL    <span class="hljs-params">| NULL |    <span class="hljs-number">2 <span class="hljs-params">| NULL  | +----+-------------+-------+------+---------------+------+---------+------+------+-------+ <span class="hljs-number">1 row <span class="hljs-keyword">in set (<span class="hljs-number">0.<span class="hljs-number">00 sec)</span></span></span></span></span></span></span></span></span></span></span></span></span></span></span></span></span>

执行计划

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60 61 62 63 64 65 66 67 68 69 70 71 72 73 74 75 76 77 78 79 80 81 82 83

id         查询顺序标识             如：mysql> explain <span class="hljs-keyword">select * <span class="hljs-keyword">from (<span class="hljs-keyword">select nid,name <span class="hljs-keyword">from tb1 <span class="hljs-keyword">where nid < <span class="hljs-number">10) <span class="hljs-keyword">as B;             +----+-------------+------------+-------+---------------+---------+---------+------+------+-------------+             | id | select_type | table      | type  | possible_keys | key     | key_len | <span class="hljs-keyword">ref  | rows | Extra       |             +----+-------------+------------+-------+---------------+---------+---------+------+------+-------------+             |  <span class="hljs-number">1 | PRIMARY     | <derived2> | ALL   | NULL          | NULL    | NULL    | NULL |    <span class="hljs-number">9 | NULL        |             |  <span class="hljs-number">2 | DERIVED     | tb1        | range | PRIMARY       | PRIMARY | <span class="hljs-number">8       | NULL |    <span class="hljs-number">9 | Using <span class="hljs-keyword">where |             +----+-------------+------------+-------+---------------+---------+---------+------+------+-------------+         特别的：如果使用union连接气值可能为<span class="hljs-literal">null       select_type         查询类型             SIMPLE          简单查询             PRIMARY         最外层查询             SUBQUERY        映射为子查询             DERIVED         子查询             UNION           联合             UNION RESULT    使用联合的结果             ...     table         正在访问的表名       type         查询时的访问方式，性能：all < index < range < index_merge < ref_or_null < <span class="hljs-keyword">ref < eq_ref < system/<span class="hljs-keyword">const             ALL             全表扫描，对于数据表从头到尾找一遍                             <span class="hljs-keyword">select * <span class="hljs-keyword">from tb1;                             特别的：如果有limit限制，则找到之后就不在继续向下扫描                                    <span class="hljs-keyword">select * <span class="hljs-keyword">from tb1 <span class="hljs-keyword">where email = <span class="hljs-string">'seven@live.com'                                    <span class="hljs-keyword">select * <span class="hljs-keyword">from tb1 <span class="hljs-keyword">where email = <span class="hljs-string">'seven@live.com' limit <span class="hljs-number">1;                                    虽然上述两个语句都会进行全表扫描，第二句使用了limit，则找到一个后就不再继续扫描。               INDEX           全索引扫描，对索引从头到尾找一遍                             <span class="hljs-keyword">select nid <span class="hljs-keyword">from tb1;               RANGE          对索引列进行范围查找                             <span class="hljs-keyword">select *  <span class="hljs-keyword">from tb1 <span class="hljs-keyword">where name < <span class="hljs-string">'alex';                             PS:                                 between and                                 <span class="hljs-keyword">in                                 >   >=  <   <=  操作                                 注意：!= 和 > 符号               INDEX_MERGE     合并索引，使用多个单列索引搜索                             <span class="hljs-keyword">select *  <span class="hljs-keyword">from tb1 <span class="hljs-keyword">where name = <span class="hljs-string">'alex' <span class="hljs-function">or nid <span class="hljs-title">in (<span class="hljs-params"><span class="hljs-number">11,<span class="hljs-number">22,<span class="hljs-number">33);               REF             根据索引查找一个或多个值                             <span class="hljs-keyword">select *  <span class="hljs-keyword">from tb1 <span class="hljs-keyword">where name = <span class="hljs-string">'seven';               EQ_REF          连接时使用primary key 或 unique类型                             <span class="hljs-keyword">select tb2.nid,tb1.name <span class="hljs-keyword">from tb2 left <span class="hljs-keyword">join tb1 on tb2.nid = tb1.nid;               CONST           常量                             表最多有一个匹配行,因为仅有一行,在这行的列值可被优化器剩余部分认为是常数,<span class="hljs-keyword">const表很快,因为它们只读取一次。                             <span class="hljs-keyword">select nid <span class="hljs-keyword">from tb1 <span class="hljs-keyword">where nid = <span class="hljs-number">2 ;               SYSTEM          系统                             表仅有一行(=系统表)。这是<span class="hljs-keyword">const联接类型的一个特例。                             <span class="hljs-keyword">select * <span class="hljs-keyword">from (<span class="hljs-keyword">select nid <span class="hljs-keyword">from tb1 <span class="hljs-keyword">where nid = <span class="hljs-number">1) <span class="hljs-keyword">as A;     possible_keys         可能使用的索引       key         真实使用的       key_len         MySQL中使用索引字节长度       rows         mysql估计为了找到所需的行而要读取的行数 ------ 只是预估值       extra         该列包含MySQL解决查询的详细信息         “Using index”             此值表示mysql将使用覆盖索引，以避免访问表。不要把覆盖索引和index访问类型弄混了。         “Using <span class="hljs-keyword">where”             这意味着mysql服务器将在存储引擎检索行后再进行过滤，许多<span class="hljs-keyword">where条件里涉及索引中的列，当（并且如果）它读取索引时，就能被存储引擎检验，因此不是所有带<span class="hljs-keyword">where子句的查询都会显示“Using <span class="hljs-keyword">where”。有时“Using <span class="hljs-keyword">where”的出现就是一个暗示：查询可受益于不同的索引。         “Using temporary”             这意味着mysql在对查询结果排序时会使用一个临时表。         “Using filesort”             这意味着mysql会对结果使用一个外部索引排序，而不是按索引次序从表里读取行。mysql有两种文件排序算法，这两种排序方式都可以在内存或者磁盘上完成，explain不会告诉你mysql将使用哪一种文件排序，也不会告诉你排序会在内存里还是磁盘上完成。         “<span class="hljs-function">Range <span class="hljs-keyword">checked <span class="hljs-keyword">for each <span class="hljs-title">record(<span class="hljs-params">index map: N)”             这个意味着没有好用的索引，新的索引将在联接的每一行上重新估算，N是显示在possible_keys列中索引的位图，并且是冗余的。</span></span></span></span></span></span></span></span></span></span></span></span></span></span></span></span></span></span></span></span></span></span></span></span></span></span></span></span></span></span></span></span></span></span></span></span></span></span></span></span></span></span></span></span></span></span></span></span></span></span></span></span></span></span></span></span></span></span></span></span></span></span></span></span></span></span></span></span></span></span></span></span></span></span></span>

慢日志查询

a、配置MySQL自动记录慢日志

1 2 3 4

<span class="hljs-attr">slow_query_log = <span class="hljs-literal">OFF                            是否开启慢日志记录 <span class="hljs-attr">long_query_time = <span class="hljs-number">2                              时间限制，超过此时间，则记录 <span class="hljs-attr">slow_query_log_file = /usr/slow.log        日志文件 <span class="hljs-attr">log_queries_not_using_indexes = <span class="hljs-literal">OFF     为使用索引的搜索是否记录</span></span></span></span></span></span></span>

注：查看当前配置信息：
　　     show variables like ‘%query%’
     修改当前配置：
　　　　set global 变量名 = 值

b、查看MySQL慢日志

mysqldumpslow -s at -a  /usr/local/var/mysql/MacBook-Pro-3-slow.log

参数详解

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26

""" <span class="hljs-comment">--verbose    版本 <span class="hljs-comment">--debug      调试 <span class="hljs-comment">--help       帮助   -v           版本 -d           调试模式 -s ORDER     排序方式              what to sort by (al, at, ar, c, l, r, t), 'at' is default               al: average <span class="hljs-keyword">lock <span class="hljs-keyword">time               ar: average <span class="hljs-keyword">rows sent               <span class="hljs-keyword">at: average <span class="hljs-keyword">query <span class="hljs-keyword">time                c: <span class="hljs-keyword">count                l: <span class="hljs-keyword">lock <span class="hljs-keyword">time                r: <span class="hljs-keyword">rows sent                t: <span class="hljs-keyword">query <span class="hljs-keyword">time -r           反转顺序，默认文件倒序拍。<span class="hljs-keyword">reverse the <span class="hljs-keyword">sort <span class="hljs-keyword">order (largest <span class="hljs-keyword">last instead <span class="hljs-keyword">of <span class="hljs-keyword">first) -t <span class="hljs-keyword">NUM       显示前N条just <span class="hljs-keyword">show the top n queries -a           不要将<span class="hljs-keyword">SQL中数字转换成N，字符串转换成S。don<span class="hljs-string">'t abstract all numbers to N and strings to 'S<span class="hljs-string">' -n NUM       abstract numbers with at least n digits within names -g PATTERN   正则匹配；grep: only consider stmts that include this string -h HOSTNAME  mysql机器名或者IP；hostname of db server for *-slow.log filename (can be wildcard),              default is '*<span class="hljs-string">', i.e. match all -i NAME      name of server instance (if using mysql.server startup script) -l           总时间中不减去锁定时间；don't subtract <span class="hljs-keyword">lock <span class="hljs-keyword">time <span class="hljs-keyword">from total <span class="hljs-keyword">time <span class="hljs-string">"""</span></span></span></span></span></span></span></span></span></span></span></span></span></span></span></span></span></span></span></span></span></span></span></span></span></span></span></span></span></span></span></span>