《Mysql学习简单谈谈MySQL的loose index scan》要点:
本文介绍了Mysql学习简单谈谈MySQL的loose index scan,希望对您有用。如果有疑问,可以联系我们。
MYSQL数据库众所周知,InnoDB采用IOT(index organization table)即所谓的索引组织表,而叶子节点也就存放了所有的数据,这就意味着,数据总是依照某种顺序存储的.所以问题来了,如果是这样一个语句,执行起来应该是怎么样的呢?语句如下:
MYSQL数据库
select count(distinct a) from table1;
MYSQL数据库 列a上有一个索引,那么依照简单的想法来讲,如何扫描呢?很简单,一条一条的扫描,这样一来,其实做了一次索引全扫描,效率很差.这种扫描方式会扫描到很多很多的重复的索引,这样说的话优化的办法也是很容易想到的:跳过重复的索引就可以了.于是网上能搜到这样的一个优化的办法:
MYSQL数据库
select count(*) from (select distinct a from table1) t;
MYSQL数据库 从已经搜索到的资料看,这样的执行计划中的extra就从using index变成了using index for group-by.
MYSQL数据库 但是,但是,但是,好在我们现在已经没有使用5.1的版本了,大家基本上都是5.5以上了,这些现代版本,已经实现了loose index scan:
MYSQL数据库 很好很好,就不需要再用这种奇技淫巧去优化SQL了.
MYSQL数据库 文档里关于group by这里写的有点意思,说是最大众化的方法就是进行全表扫描并且创建一个临时表,这样执行计划就会难看的要命了,肯定有ALL和using temporary table了.
MYSQL数据库5.0之后group by在特定条件下可能使用到loose index scan,
MYSQL数据库
CREATE TABLE log_table (
id INT NOT NULL PRIMARY KEY,
log_machine VARCHAR(20) NOT NULL,
log_time DATETIME NOT NULL
) ENGINE=InnoDB DEFAULT CHARSET=utf8;
CREATE INDEX ix_log_machine_time ON log_table (log_machine, log_time);
MYSQL数据库1
MYSQL数据库
SELECT MAX(log_time) FROM log_table;
SELECT MAX(log_time) FROM log_table WHERE log_machine IN ('Machine 1');
MYSQL数据库这两条sql都只需一次index seek便可返回,源于索引的有序排序,优化器意识到min/max位于最左/右块,从而避免范围扫描;
extra显示Select tables optimized away ;
2
MYSQL数据库执行计划type 为range(extra显示using where; using index),即执行索引范围扫描,先读取所有满足log_machine约束的记录,然后对其遍历找出max value;
改进
MYSQL数据库
SELECT MAX(log_time) FROM log_table WHERE log_machine IN (‘Machine 1')
Union
SELECT MAX(log_time) FROM log_table WHERE log_machine IN (‘Machine 2')
…..
MYSQL数据库即对每个log_machine执行loose index scan,rows从原来的82636下降为16(该表总共1,000,000条记录).
MYSQL数据库Group by何时使用loose index scan?
MYSQL数据库适用条件:
MYSQL数据库1 针对单表操作
2 Group by使用索引的最左前缀列
3 只支持聚集函数min()/max()
4 Where条件出现的列必须为=constant操作 , 没出现在group by中的索引列必须使用constant
5 不支持前缀索引,即部分列索引 ,如index(c1(10))
执行计划的extra应该显示using index for group-by
假定表t1有个索引idx(c1,c2,c3)
MYSQL数据库
SELECT c1, c2 FROM t1 GROUP BY c1, c2;
SELECT DISTINCT c1, c2 FROM t1;
SELECT c1, MIN(c2) FROM t1 GROUP BY c1;
SELECT c1, c2 FROM t1 WHERE c1 < const GROUP BY c1, c2;
SELECT MAX(c3), MIN(c3), c1, c2 FROM t1 WHERE c2 > const GROUP BY c1, c2;
SELECT c2 FROM t1 WHERE c1 < const GROUP BY c1, c2;
SELECT c1, c2 FROM t1 WHERE c3 = const GROUP BY c1, c2
SELECT c1, c3 FROM t1 GROUP BY c1, c2;--无法使用松散索引
MYSQL数据库而SELECT c1, c3 FROM t1 where c3= const GROUP BY c1, c2;则可以
MYSQL数据库紧凑索引扫描tight index scan
Group by在无法使用loose index scan,还可以选择tight,若两者都不可选,则只能借助临时表;
扫描索引时,须读取所有满足条件的索引键,要么是全索引扫描,要么是范围索引扫描;
Group by的索引列不连续;或者不是从最左前缀开始,但是where条件里出现最左列;
MYSQL数据库
SELECT c1, c2, c3 FROM t1 WHERE c2 = 'a' GROUP BY c1, c3;
SELECT c1, c2, c3 FROM t1 WHERE c1 = 'a' GROUP BY c2, c3;
MYSQL数据库5.6的改进
事实上,5.6的index condition push down可以弥补loose index scan缺失带来的性能损失.
KEY(age,zip)
MYSQL数据库
mysql> explain SELECT name FROM people WHERE age BETWEEN 18 AND 20 AND zip IN (12345,12346, 12347);
+----+-------------+--------+-------+---------------+------+---------+------+-------+-------------+
| id | select_type | table | type | possible_keys | key | key_len | ref | rows | Extra |
+----+-------------+--------+-------+---------------+------+---------+------+-------+-------------+
| 1 | SIMPLE | people | range | age | age | 4 | NULL | 90556 | Using where |
+----+-------------+--------+-------+---------------+------+---------+------+-------+-------------+
1 row in set (0.01 sec)
MYSQL数据库根据key_len=4可以推测出sql只用到索引的第一列,即先通过索引查出满足age (18,20)的行记录,然后从server层筛选出满足zip约束的行;
pre-5.6,对于复合索引,只有当引导列使用"="时才有机会在索引扫描时使用到后面的索引列.
MYSQL数据库
mysql> explain SELECT name FROM people WHERE age=18 AND zip IN (12345,12346, 12347);
+----+-------------+--------+-------+---------------+------+---------+------+------+-------------+
| id | select_type | table | type | possible_keys | key | key_len | ref | rows | Extra |
+----+-------------+--------+-------+---------------+------+---------+------+------+-------------+
| 1 | SIMPLE | people | range | age | age | 8 | NULL | 3 | Using where |
+----+-------------+--------+-------+---------------+------+---------+------+------+-------------+
1 row in set (0.00 sec)
MYSQL数据库对比一下查询效率
MYSQL数据库
mysql> SELECT sql_no_cache name FROM people WHERE age=19 AND zip IN (12345,12346, 12347);
+----------------------------------+
| name |
+----------------------------------+
| 888ba838661aff00bbbce114a2a22423 |
+----------------------------------+
1 row in set (0.06 sec)
mysql> SELECT SQL_NO_CACHE name FROM people WHERE age BETWEEN 18 AND 22 AND zip IN (12345,12346, 12347);
+----------------------------------+
| name |
+----------------------------------+
| ed4481336eb9adca222fd404fa15658e |
| 888ba838661aff00bbbce114a2a22423 |
+----------------------------------+
2 rows in set (1 min 56.09 sec)
MYSQL数据库对于第二条sql,可以使用union改写,
MYSQL数据库
mysql> SELECT name FROM people WHERE age=18 AND zip IN (12345,12346, 12347)
-> UNION ALL
-> SELECT name FROM people WHERE age=19 AND zip IN (12345,12346, 12347)
-> UNION ALL
-> SELECT name FROM people WHERE age=20 AND zip IN (12345,12346, 12347)
-> UNION ALL
-> SELECT name FROM people WHERE age=21 AND zip IN (12345,12346, 12347)
-> UNION ALL
-> SELECT name FROM people WHERE age=22 AND zip IN (12345,12346, 12347);
MYSQL数据库而mysql5.6引入了index condition pushdown,从优化器层面办理了此类问题.
维易PHP培训学院每天发布《Mysql学习简单谈谈MySQL的loose index scan》等实战技能,PHP、MYSQL、LINUX、APP、JS,CSS全面培养人才。