猿问

MySQL 关于查询时扫描行数与索引的疑问

测试表:
CREATETABLE`table_1`(
`id`int(10)unsignedNOTNULLAUTO_INCREMENT,
`title`textNOTNULL,
`category_id`int(10)unsignedNOTNULL,
PRIMARYKEY(`id`)
)ENGINE=InnoDB;
其中id字段是自增主键
插入30行用于测试的数据:
insertintotable_1(`category_id`)values(1);
insertintotable_1(`category_id`)values(1);
insertintotable_1(`category_id`)values(1);
insertintotable_1(`category_id`)values(2);
insertintotable_1(`category_id`)values(2);
insertintotable_1(`category_id`)values(2);
insertintotable_1(`category_id`)values(2);
insertintotable_1(`category_id`)values(2);
insertintotable_1(`category_id`)values(2);
insertintotable_1(`category_id`)values(2);
insertintotable_1(`category_id`)values(2);
insertintotable_1(`category_id`)values(2);
insertintotable_1(`category_id`)values(2);
insertintotable_1(`category_id`)values(2);
insertintotable_1(`category_id`)values(2);
insertintotable_1(`category_id`)values(2);
insertintotable_1(`category_id`)values(2);
insertintotable_1(`category_id`)values(2);
insertintotable_1(`category_id`)values(2);
insertintotable_1(`category_id`)values(2);
insertintotable_1(`category_id`)values(2);
insertintotable_1(`category_id`)values(3);
insertintotable_1(`category_id`)values(3);
insertintotable_1(`category_id`)values(3);
insertintotable_1(`category_id`)values(4);
insertintotable_1(`category_id`)values(4);
insertintotable_1(`category_id`)values(4);
insertintotable_1(`category_id`)values(5);
insertintotable_1(`category_id`)values(5);
insertintotable_1(`category_id`)values(5);
执行查询:
mysql>explainselect*from`table_1`orderby`id`DESClimit0,5;
+----+-------------+---------+-------+---------------+---------+---------+------+------+-------+
|id|select_type|table|type|possible_keys|key|key_len|ref|rows|Extra|
+----+-------------+---------+-------+---------------+---------+---------+------+------+-------+
|1|SIMPLE|table_1|index|NULL|PRIMARY|4|NULL|5||
+----+-------------+---------+-------+---------------+---------+---------+------+------+-------+
1rowinset
这个很好理解,因为id是主键,查询中只使用了orderbyid,查询涉及记录行数rows5,因为limit0,5
mysql>explainselect*from`table_1`where`category_id`=2orderby`id`DESClimit0,5;
+----+-------------+---------+-------+---------------+---------+---------+------+------+-------------+
|id|select_type|table|type|possible_keys|key|key_len|ref|rows|Extra|
+----+-------------+---------+-------+---------------+---------+---------+------+------+-------------+
|1|SIMPLE|table_1|index|NULL|PRIMARY|4|NULL|5|Usingwhere|
+----+-------------+---------+-------+---------------+---------+---------+------+------+-------------+
1rowinset
这个就无法理解了,为什么使用了wherecategory_id=2,用一个非索引字段where,该查询涉及的记录数仍然是5?将category_id=2改为任何数字,rows都为5,实际记录前几条并不是category_id=2,按理应该先跳过category_id!=2的然后筛选出符合的结果返回,这样涉及的行数应该大于5啊
更无法理解的是,如果使用该表category_id建立索引,同样该SQL执行结果:
mysql>explainselect*from`table_1`where`category_id`=2orderby`id`DESClimit0,5;
+----+-------------+---------+------+---------------+-------------+---------+-------+------+-------------+
|id|select_type|table|type|possible_keys|key|key_len|ref|rows|Extra|
+----+-------------+---------+------+---------------+-------------+---------+-------+------+-------------+
|1|SIMPLE|table_1|ref|category_id|category_id|4|const|18|Usingwhere|
+----+-------------+---------+------+---------------+-------------+---------+-------+------+-------------+
1rowinset
也就是wherecategory_id=2涉及行数成了category_id=2记录的总数!也就是18条
那么如果数据库中有1千万条数据,均分至category_id1-10的话,这时候需要执行:
select*from`table_1`where`category_id`=2orderby`id`DESClimit0,5;
是否需要建立category_id索引呢?如果建立每次都要扫描100万条索引记录吗?如果不建立任何索引,该SQL是否会存在性能问题?
慕无忌1623718
浏览 1512回答 2
2回答

米琪卡哇伊

《高性能mysql》第三版p698ROWS列这一列是Mysql估计为了找到所需的行而要读取的行数。......根据表的统计信息和索引的选用情况,这个估算值可能很不准确。由于你的category_id没有索引,mysql认为他可能要把每一条都遍历一次,才可以找到,所以rows是5(因为是估计的)。category_id要不要加索引,我觉得和你category_id的数量有关。如果category_id就只有两条的话(比如sex性别,只有男和女),那么加索引完全是浪费资源,如果category过多,那就加上吧。

侃侃无极

第一个问题为什么还是5呢,因为explain后面的sql并没有真正执行,mysql只是根据这条sql预测的,所以这个值肯定是不准确的估计值,而此时mysql预估的值就是limit的5,最好情况5行就够,因为sql没有执行,mysql没法给你预估出别的值,只能是sql里存在的值。第二个问题category_id建立索引后预估的值变成了category_id=2的行数,这是因为mysql执行sql会按照sql中索引的顺序来使用,即这条sql会使用category_id的索引,而不会使用id的索引(新版本mysql可能两个索引都会使用),通过category_id的索引得到所有category_id=2的行后要整体进行orderby,所以预估的值就是category_id=2的的行数,因为前面说过explain不会真正执行sql,所以category_id=2的值应该是存在索引中的(猜的)。第三个问题,通过前面说的,可以了解到,mysql不会真正扫描100万条索引记录的(当然如果mysql不能使用orderby的索引还是要扫描100万条索引记录的),你可以先不建立索引,如果性能不够再建立索引,这要求你的mysql可以在线DDL
随时随地看视频慕课网APP

相关分类

JavaScript
我要回答