数据库调优分享------参考一本mysql资料书
日常的困扰
**反映查询某个数据越来越慢，怎么办？
数据库监控页面显示某个操作频繁执行，每次执行都非常耗时，怎么办？
1.Sql优化
2.合理使用索引
一、sql优化—常用方法
1）not in 子查询优化
2）模糊匹配 like ‘%abc%’
3）limit分页优化
4）count(*)统计数据如何加快速度
5）or条件如何优化
6）用where子句代替having子句
1、not in 子查询优化
SELECT count(*) FROM t_cps t1 where t1.flag>=2 and sample_md5 not in (select sample_hash from t_sfa_sample t2 where t2.sample_hash=t1.sample_md5);
-1052661行记录
33.530s
SELECT count(*) FROM t_cps t1 where t1.flag>=2 and not exists(select sample_hash from t_sfa_sample t2 where t2.sample_hash=t1.sample_md5);
-1052661行记录
18.617s
SELECT count(*) FROM t_cps t1 left join t_sfa_sample t2 on t1.sample_md5=t2.sample_hash where t1.flag>=2 and t2.sample_hash is null;
-1052661行记录
14.544s
生产环境中，尽量避免使用子查询，可用表连接join代替。可避免mysql在内存中创建临时表
in和exists
SELECT count(*) FROM t_cps t1 where t1.flag>=2 and sample_md5 in (select sample_hash from t_sfa_sample t2 where t2.sample_hash=t1.sample_md5);
1227992行记录
30.762s
SELECT count(*) FROM t_cps t1 where t1.flag>=2 and exists (select sample_hash from t_sfa_sample t2 where t2.sample_hash=t1.sample_md5);
1227992行记录
15.581s
in和exists执行时:
in是先执行子查询中的查询，然后再执行主查询。
exists是先执行主查询，即外层表的查询，然后再执行子查询
exists 和 in 在执行时效率单从执行时间来说差不多，exists要稍微优于in。在使用时一般应该是用exists而不用in
而在not in 和 not exists比较时，not exists的效率要比not in的效率要高。
但也有少数时候并非如此。
建议是尽量多使用exists和not exists，少用in 和not in
2、模糊匹配 like ‘%abc%’
mysql> explain select * from update_log_test where msg like 'e%';
+----+-------------+-----------------+-------+---------------+---------+---------+------+------+-----------------------+
| id | select_type | table | type | possible_keys | key | key_len | ref | rows | Extra |
+----+-------------+-----------------+-------+---------------+---------+---------+------+------+-----------------------+
| 1 | SIMPLE | update_log_test | range | idx_msg | idx_msg | 768 | NULL | 1 | Using index condition |
+----+-------------+-----------------+-------+---------------+---------+---------+------+------+-----------------------+
1 row in set (0.00 sec)
mysql> explain select * from update_log_test where msg like '%e%';
+----+-------------+-----------------+------+---------------+------+---------+------+------+-------------+
| id | select_type | table | type | possible_keys | key | key_len | ref | rows | Extra |
+----+-------------+-----------------+------+---------------+------+---------+------+------+-------------+
| 1 | SIMPLE | update_log_test | ALL | NULL | NULL | NULL | NULL | 4 | Using where |
+----+-------------+-----------------+------+---------------+------+---------+------+------+-------------+
1 row in set (0.00 sec)
like ‘abc%’可以用到索引，但like‘%abc%’却不行。
SELECT * FROM books WHERE name like "MySQL%" ;
但是如果换用下面的查询，返回的结果一样，但速度就要快上很多：
SELECT * FROM books WHERE name>="MySQL"and name<"MySQM" ;
补充：explain列的解释
table 显示这一行的数据是关于哪张表的
type 这是重要的列，显示连接使用了何种类型。从最好到最差的连接类型为const、eq_reg、ref、range、index和ALL
possible_keys 显示可能应用在这张表中的索引。如果为空，没有可能的索引。
key 实际使用的索引。如果为NULL，则没有使用索引。很少的情况下，MYSQL会选择优化不足的索引。这种情况下，可以在SELECT语句中使用USE INDEX（indexname）来强制使用一个索引或者用IGNORE INDEX（indexname）来强制MYSQL忽略索引
key_len 使用的索引的长度。在不损失精确性的情况下，长度越短越好
ref 显示索引的哪一列被使用了，显示了哪些字段或者常量被用来和 key配合从表中查询记录出来。
rows MYSQL认为必须检查的用来返回请求数据的行数
Extra 关于MYSQL如何解析查询的额外信息。这里可以看到的坏的例子是Using temporary和Using filesort，意思MYSQL根本不能使用索引，检索会很慢
type不同连接类型的解释（按照效率高低的顺序排序）
const 索引可以是主键或惟一索引，因为只有一行，这个值实际就是常数，因为MYSQL先读这个值然后把它当做常数来对待
mysql> explain select * from update_log_20141111 where id=2;
+----+-------------+---------------------+-------+---------------+---------+---------+-------+------+-------+
| id | select_type | table | type | possible_keys | key | key_len | ref | rows | Extra |
+----+-------------+---------------------+-------+---------------+---------+---------+-------+------+-------+
| 1 | SIMPLE | update_log_20141111 | const | PRIMARY | PRIMARY | 4 | const | 1 | NULL |
+----+-------------+---------------------+-------+---------------+---------+---------+-------+------+-------+
eq_ref 从前面的表中，对每一个记录的联合都从表中读取一个记录，它在查询使用了索引为主键或惟一键的全部时使用
mysql> explain select * from t1,t2 where t1.id=t2.id;
+----+-------------+-------+--------+---------------+---------+---------+------------+------+-------+
| id | select_type | table | type | possible_keys | key | key_len | ref | rows | Extra |
+----+-------------+-------+--------+---------------+---------+---------+------------+------+-------+
| 1 | SIMPLE | t2 | ALL | PRIMARY | NULL | NULL | NULL | 9 | NULL |
| 1 | SIMPLE | t1 | eq_ref | PRIMARY | PRIMARY | 4 | test.t2.id | 1 | NULL |
+----+-------------+-------+--------+---------------+---------+---------+------------+------+-------+
2 rows in set (0.00 sec)
ref 只有在查询使用了不是惟一或主键的键或者是这些类型的部分（比如，利用最左边前缀）时发生。对于之前的表的每一个行联合，全部记录都将从表中读出。这个类型严重依赖于根据索引匹配的记录多少—越少越好
mysql> explain select * from update_log_20141110 t1,update_log_test t2 where t1.action=t2.action;
+----+-------------+-------+------+---------------+----------+---------+----------------+------+-------------+
| id | select_type | table | type | possible_keys | key | key_len | ref | rows | Extra |
+----+-------------+-------+------+---------------+----------+---------+----------------+------+-------------+
| 1 | SIMPLE | t1 | ALL | i_action | NULL | NULL | NULL | 2 | Using where |
| 1 | SIMPLE | t2 | ref | i_action | i_action | 2 | test.t1.action | 1 | NULL |
+----+-------------+-------+------+---------------+----------+---------+----------------+------+-------------+
2 rows in set (0.00 sec)
range 返回一个范围中的行，比如使用>或<查找东西时
mysql> explain select * from t1 where id>=9900 order by id limit 10;
+----+-------------+-------+-------+---------------+---------+---------+------+------+-------------+
| id | select_type | table | type | possible_keys | key | key_len | ref | rows | Extra |
+----+-------------+-------+-------+---------------+---------+---------+------+------+-------------+
| 1 | SIMPLE | t1 | range | PRIMARY | PRIMARY | 4 | NULL | 101 | Using where |
+----+-------------+-------+-------+---------------+---------+---------+------+------+-------------+
1 row in set (0.00 sec)
index 对前面的表中的每一个记录联合进行完全扫描（比ALL更好，因为索引一般小于表数据）
ALL 对每一个记录联合进行完全扫描，这一般比较糟糕，应该尽量避免
Extra 列返回的描述的意义
Distinct 一旦MYSQL找到了与行相联合匹配的行，就不再搜索了
Not exists MYSQL优化了LEFT JOIN，一旦它找到了匹配LEFT JOIN标准的行，就不再搜索
Record（index map:#）没有找到理想的索引，因此对于从前面表中来的每一个行组合，MYSQL检查使用哪个索引，并用它来从表中返回行。这是使用索引的最慢的连接之一
Using filesort 看到这个的时候，查询就需要优化了。MYSQL需要进行额外的步骤来发现如何对返回的行排序。它根据连接类型以及存储排序键值和匹配条件的全部行的行指针来排序全部行
Using index 列数据是从仅仅使用了索引中的信息而没有读取实际的行动的表返回的，这发生在对表的全部的请求列都是同一个索引的部分的时候
Using temporary 看到这个的时候，查询需要优化了。这里，MYSQL需要创建一个临时表来存储结果，这通常发生在对不同的列集进行ORDER BY上，而不是GROUP BY上
Using where 使用了WHERE从句来限制哪些行将与下一张表匹配或者是返回给用户。如果不想返回表中的全部行，并且连接类型ALL或index，这就会发生，或者是查询有问题
3、limit分页优化
Select * from t1 order by id limit 9900,10;
上面的语句，虽然用到了id索引，但是从第一行开始起定位至9900行，然后再扫描后10行，相当于进行了一次全扫描，显然效率不高。
Select * from t1 where id>=9900 order by id limit 10;
利用id索引直接定位到9900行，然后在扫描出后10行，相当于一个range范围扫描
mysql> explain select * from t1 order by id limit 9900,10;
+----+-------------+-------+-------+---------------+---------+---------+------+------+-------+
| id | select_type | table | type | possible_keys | key | key_len | ref | rows | Extra |
+----+-------------+-------+-------+---------------+---------+---------+------+------+-------+
| 1 | SIMPLE | t1 | index | NULL | PRIMARY | 4 | NULL | 9910 | NULL |
+----+-------------+-------+-------+---------------+---------+---------+------+------+-------+
1 row in set (0.00 sec)
mysql> explain select * from t1 where id>=9900 order by id limit 10;
+----+-------------+-------+-------+---------------+---------+---------+------+------+-------------+
| id | select_type | table | type | possible_keys | key | key_len | ref | rows | Extra |
+----+-------------+-------+-------+---------------+---------+---------+------+------+-------------+
| 1 | SIMPLE | t1 | range | PRIMARY | PRIMARY | 4 | NULL | 101 | Using where |
+----+-------------+-------+-------+---------------+---------+---------+------+------+-------------+
1 row in set (0.00 sec)
4、count(*)统计数据如何加快速度
select count(*) from update_log;
select count(*) from update_log where sid>=0; 利用辅助索引
select count(distinct k) from t;
select count(*) from (select distinct k from t) tmp; 利用索引来做排重操作。
注意：innodb count（*）必须全表扫，而不像myisam那样有一个计数器，直接从中取出数据。Innodb必须要全表扫一次才能得到count，而且会锁表。
5、or条件如何优化
select * from user where name='d' or age=31;
name和age都建立了索引，但explain发现这个or条件用不到索引。
改成union all结果集合并
select * from user where name='d' union all select * from user where age=31;
6、用where子句代替having子句
select * from user group by id having id>40 limit 3;
select * from user where id>40 group by id limit 3;
having只会在检索出所有记录之后才对结果集进行过滤。这个处理需要排序、总计等操作。
如果能通过where子句限制记录的数目，那就能减少这方面的开销。
SQL优化的一般步骤
1、通过 show status 命令了解各种SQL的执行频率
show status like 'uptime' 当前MySQL运行时间
show status like 'com_select' 当前MySQL执行了多少次查询
show status like 'com_insert' 当前MySQL执行了多少次添加
show status like 'com_update' 当前MySQL执行了多少次更新
show status like 'com_delete' 当前MySQL执行了多少次删除
show status 语法：
show [session|global] status like '';
如果不写 [session|global] 表示默认是 session 指取出当前窗口的执行情况
如果想看所有(mysql启动到现在)的情况 加上 global
show global status like 'com_insert';
2、 定位执行效率较低的SQL语句 （重点select）
3、通过 explain 分析低效率的SQL语句的执行情况
4、确定问题并采取相应的优化措施
二、合理使用索引
使用索引，why?
单列索引和联合索引
字段使用函数，将不能走索引
当取出的数据量超过表中数据的20%，索引是否有效？
order by 和group by优化
全文索引
1、适当的索引对应用的性能来说至关重要。
2、索引只对select有加速作用，但对写入（insert,update、delete）操作会带来额外的开销，如果带有一个或多个索引，那么数据更新的时候，mysql也要更新各个索引。
3、并不是所有经常查询的列都适合创建索引，区分度不高的，通常走全表扫浏览会更快。例如性别，只有男女两种，就不适合。
4、一条sql只能用一个索引，如果有多个，优化器会选择最优的。
1、单列索引和联合索引
mysql> create index i_s_time on update_log(server,time);
联合索引要遵循最左侧原则
mysql> explain select * from update_log where server='115.29.138.24' and time='13:00:00';
+----+-------------+------------+------+-----------------+----------+---------+-------------+------+-----------------------+
| id | select_type | table | type | possible_keys | key | key_len | ref | rows | Extra |
+----+-------------+------------+------+-----------------+----------+---------+-------------+------+-----------------------+
| 1 | SIMPLE | update_log | ref | i_time,i_s_time | i_s_time | 50 | const,const | 7 | Using index condition |
+----+-------------+------------+------+-----------------+----------+---------+-------------+------+-----------------------+
1 row in set (0.00 sec)
mysql> explain select * from update_log where server='115.29.138.24';
+----+-------------+------------+------+---------------+----------+---------+-------+---------+-----------------------+
| id | select_type | table | type | possible_keys | key | key_len | ref | rows | Extra |
+----+-------------+------------+------+---------------+----------+---------+-------+---------+-----------------------+
| 1 | SIMPLE | update_log | ref | i_s_time | i_s_time | 46 | const | 2653264 | Using index condition |
+----+-------------+------------+------+---------------+----------+---------+-------+---------+-----------------------+
1 row in set (0.00 sec)
mysql> explain select * from update_log where server='115.29.138.24' and client='14.197.74.21' and time='13:00:00';
+----+-------------+------------+------+-----------------+----------+---------+-------------+------+------------------------------------+
| id | select_type | table | type | possible_keys | key | key_len | ref | rows | Extra |
+----+-------------+------------+------+-----------------+----------+---------+-------------+------+------------------------------------+
| 1 | SIMPLE | update_log | ref | i_time,i_s_time | i_s_time | 50 | const,const | 7 | Using index condition; Using where |
+----+-------------+------------+------+-----------------+----------+---------+-------------+------+------------------------------------+
1 row in set (0.00 sec)
mysql> explain select * from update_log where time='13:00:00';
+----+-------------+------------+------+---------------+------+---------+------+----------+-------------+
| id | select_type | table | type | possible_keys | key | key_len | ref | rows | Extra |
+----+-------------+------------+------+---------------+------+---------+------+----------+-------------+
| 1 | SIMPLE | update_log | ALL | NULL | NULL | NULL | NULL | 21615394 | Using where |
+----+-------------+------------+------+---------------+------+---------+------+----------+-------------+
1 row in set (0.00 sec)
2、字段使用函数，将不能走索引 --mysql目前还不支持函数索引
mysql> explain select * from update_log where year(date)<2014;
+----+-------------+------------+------+---------------+------+---------+------+----------+-------------+
| id | select_type | table | type | possible_keys | key | key_len | ref | rows | Extra |
+----+-------------+------------+------+---------------+------+---------+------+----------+-------------+
| 1 | SIMPLE | update_log | ALL | NULL | NULL | NULL | NULL | 21615394 | Using where |
+----+-------------+------------+------+---------------+------+---------+------+----------+-------------+
1 row in set (0.00 sec)
在where后面的查询条件字段使用了date()函数，是不会用到索引的。
mysql> explain select * from update_log where date<'2014-01-01';
+----+-------------+------------+-------+---------------+------+---------+------+------+-----------------------+
| id | select_type | table | type | possible_keys | key | key_len | ref | rows | Extra |
+----+-------------+------------+-------+---------------+------+---------+------+------+-----------------------+
| 1 | SIMPLE | update_log | range | date | date | 4 | NULL | 1 | Using index condition |
+----+-------------+------------+-------+---------------+------+---------+------+------+-----------------------+
1 row in set (0.00 sec)
同样的情形也会发生在对数值型字段进行计算的时候：
SELECT * FROM t1 WHERE Amount/7<24;
SELECT * FROM t1 WHERE Amount<24*7;
3、无引号导致全表扫描，无法使用索引
mysql> explain select * from update_log_test where msg=123;
+----+-------------+-----------------+------+---------------+------+---------+------+------+-------------+
| id | select_type | table | type | possible_keys | key | key_len | ref | rows | Extra |
+----+-------------+-----------------+------+---------------+------+---------+------+------+-------------+
| 1 | SIMPLE | update_log_test | ALL | i_msg | NULL | NULL | NULL | 4 | Using where |
+----+-------------+-----------------+------+---------------+------+---------+------+------+-------------+
1 row in set (0.00 sec)
mysql> explain select * from update_log_test where msg='yoyo';
+----+-------------+-----------------+------+---------------+-------+---------+-------+------+-----------------------+
| id | select_type | table | type | possible_keys | key | key_len | ref | rows | Extra |
+----+-------------+-----------------+------+---------------+-------+---------+-------+------+-----------------------+
| 1 | SIMPLE | update_log_test | ref | i_msg | i_msg | 768 | const | 1 | Using index condition |
+----+-------------+-----------------+------+---------------+-------+---------+-------+------+-----------------------+
由于msg是varchar类型，因此查询的时候，必须加‘’
数字当字符类型使用时，也一定要加上。
4、当取出的数据量超过表中数据的20%，优化器认为全表扫更快，不会走索引
mysql> explain select * from update_log where time<'14:00:00';
+----+-------------+------------+------+---------------+------+---------+------+----------+-------------+
| id | select_type | table | type | possible_keys | key | key_len | ref | rows | Extra |
+----+-------------+------------+------+---------------+------+---------+------+----------+-------------+
| 1 | SIMPLE | update_log | ALL | i_time | NULL | NULL | NULL | 21615394 | Using where |
+----+-------------+------------+------+---------------+------+---------+------+----------+-------------+
1 row in set (0.00 sec)
mysql> explain select * from update_log where time<'14:00:00' and time>'13:00:00';
+----+-------------+------------+-------+---------------+--------+---------+------+---------+-----------------------+
| id | select_type | table | type | possible_keys | key | key_len | ref | rows | Extra |
+----+-------------+------------+-------+---------------+--------+---------+------+---------+-----------------------+
| 1 | SIMPLE | update_log | range | i_time | i_time | 4 | NULL | 3013332 | Using index condition |
+----+-------------+------------+-------+---------------+--------+---------+------+---------+-----------------------+
1 row in set (0.00 sec)
5、order by 和group by优化
mysql> explain select * from update_log where server='115.29.138.24' order by time;
+----+-------------+------------+------+---------------+------+---------+------+----------+-----------------------------+
| id | select_type | table | type | possible_keys | key | key_len | ref | rows | Extra |
+----+-------------+------------+------+---------------+------+---------+------+----------+-----------------------------+
| 1 | SIMPLE | update_log | ALL | NULL | NULL | NULL | NULL | 21615394 | Using where; Using filesort |
+----+-------------+------------+------+---------------+------+---------+------+----------+-----------------------------+
1 row in set (0.00 sec)
创建联合索引
mysql> create index i_s_time on update_log(server,time);
Query OK, 0 rows affected (3 min 18.94 sec)
Records: 0 Duplicates: 0 Warnings: 0
mysql> explain select * from update_log where server='115.29.138.24' order by time;
+----+-------------+------------+------+---------------+----------+---------+-------+---------+------------------------------------+
| id | select_type | table | type | possible_keys | key | key_len | ref | rows | Extra |
+----+-------------+------------+------+---------------+----------+---------+-------+---------+------------------------------------+
| 1 | SIMPLE | update_log | ref | i_s_time | i_s_time | 46 | const | 2653264 | Using index condition; Using where |
+----+-------------+------------+------+---------------+----------+---------+-------+---------+------------------------------------+
1 row in set (0.01 sec)
如果order by 后面有多个字段排序，它们的顺序要一致，如果一个是降序，一个是升序，也会出现using filesort排序。
6、全文索引
全文索引：主要是针对文件，文本的索引，比如文章(5.6开始innodb也支持）
字段类型：char、varchar、text
全文搜索通过 MATCH() 函数完成.
mysql> create table articles(
-> id int unsigned auto_increment not null primary key,
-> title varchar(200),
-> body text,
-> fulltext(title,body));
Query OK, 0 rows affected (0.05 sec)
mysql> INSERT INTO articles VALUES
-> (NULL,'MySQL Tutorial', 'DBMS stands for DataBase ...'),
-> (NULL,'How To Use MySQL Efficiently', 'After you went through a ...'),
-> (NULL,'Optimising MySQL','In this tutorial we will show ...'),
-> (NULL,'1001 MySQL Tricks','1. Never run mysqld as root. 2. ...'),
-> (NULL,'MySQL vs. YourSQL', 'In the following database comparison ...'),
-> (NULL,'MySQL Security', 'When configured properly, MySQL ...');
Query OK, 6 rows affected (0.01 sec)
Records: 6 Duplicates: 0 Warnings: 0
mysql> explain select * from articles where match(title,body) against('database');
+----+-------------+----------+----------+---------------+-------+---------+------+------+-------------+
| id | select_type | table | type | possible_keys | key | key_len | ref | rows | Extra |
+----+-------------+----------+----------+---------------+-------+---------+------+------+-------------+
| 1 | SIMPLE | articles | fulltext | title | title | 0 | NULL | 1 | Using where |
+----+-------------+----------+----------+---------------+-------+---------+------+------+-------------+
1 row in set (0.00 sec)
但如果某个单词单词出现在至少全文的50%的行中，它会被列入停止字。对于大型数据集，使用这个操作最合适不过了----一个自然语言问询不会从一个1GB 的表每隔一行返回一次。对于小型数据集，它的用处可能比较小。不是大型的数据量，就不要用他，影响插入速度
7、mysql 5.6支持explain update/delete
mysql> explain update update_log_test set msg='gugu' where id=4;
+----+-------------+-----------------+-------+---------------+---------+---------+-------+------+-------------+
| id | select_type | table | type | possible_keys | key | key_len | ref | rows | Extra |
+----+-------------+-----------------+-------+---------------+---------+---------+-------+------+-------------+
| 1 | SIMPLE | update_log_test | range | PRIMARY | PRIMARY | 4 | const | 1 | Using where |
+----+-------------+-----------------+-------+---------------+---------+---------+-------+------+-------------+
1 row in set (0.01 sec)
mysql> explain delete from update_log_test where id=4;
+----+-------------+-----------------+-------+---------------+---------+---------+-------+------+-------------+
| id | select_type | table | type | possible_keys | key | key_len | ref | rows | Extra |
+----+-------------+-----------------+-------+---------------+---------+---------+-------+------+-------------+
| 1 | SIMPLE | update_log_test | range | PRIMARY | PRIMARY | 4 | const | 1 | Using where |
+----+-------------+-----------------+-------+---------------+---------+---------+-------+------+-------------+
1 row in set (0.00 sec)
8、mysql5.6优化了合并索引
mysql> explain select * from update_log where date='2014-11-12' or time='14:00:00';
+----+-------------+------------+-------------+---------------+-------------+---------+------+----------+---------------------------------------+
| id | select_type | table | type | possible_keys | key | key_len | ref | rows | Extra |
+----+-------------+------------+-------------+---------------+-------------+---------+------+----------+---------------------------------------+
| 1 | SIMPLE | update_log | index_merge | date,i_time | date,i_time | 4,4 | NULL | 10808103 | Using union(date,i_time); Using where |
+----+-------------+------------+-------------+---------------+-------------+---------+------+----------+---------------------------------------+
1 row in set (0.01 sec)
mysql> explain select * from update_log where date='2014-11-12' union select * from update_log where time='14:00:00';
+----+--------------+------------+------+---------------+--------+---------+-------+----------+-----------------------+
| id | select_type | table | type | possible_keys | key | key_len | ref | rows | Extra |
+----+--------------+------------+------+---------------+--------+---------+-------+----------+-----------------------+
| 1 | PRIMARY | update_log | ref | date | date | 4 | const | 10807697 | NULL |
| 2 | UNION | update_log | ref | i_time | i_time | 4 | const | 406 | Using index condition |
| NULL | UNION RESULT | <union1,2> | ALL | NULL | NULL | NULL | NULL | NULL | Using temporary |
+----+--------------+------------+------+---------------+--------+---------+-------+----------+-----------------------+
3 rows in set (0.00 sec)
实际测试2条sql，or的执行时间是union执行时间的两倍。
1、union all 不一定就比 or及in 快，要结合实际情况分析到底使用哪种情况。
2、对于索引列来最好使用union all，因复杂的查询【包含运算等】将使or、in放弃索引而全表扫描，除非你能确定or、in会使用索引。
3、对于只有非索引字段来说你就老老实实的用or 或者in，因为 非索引字段本来要全表扫描而union all 只成倍增加表扫描的次数。
Union：对两个结果集进行并集操作，不包括重复行，同时进行默认规则的排序；Union All：对两个结果集进行并集操作，包括重复行，不进行排序；
union因为要进行重复值扫描，所以效率低。