Explain tool actual operation

new table ：

CREATE TABLE `actor` (
 `id` int(11) NOT NULL,
 `name` varchar(45) DEFAULT NULL,
 `update_time` datetime DEFAULT NULL,
 PRIMARY KEY (`id`)
 ) ENGINE=InnoDB DEFAULT CHARSET=utf8;

CREATE TABLE `film` (
 `id` int(11) NOT NULL AUTO_INCREMENT,
 `name` varchar(10) DEFAULT NULL,
 PRIMARY KEY (`id`),
 KEY `idx_name` (`name`)
 ) ENGINE=InnoDB DEFAULT CHARSET=utf8;

CREATE TABLE `film_actor` (
 `id` int(11) NOT NULL,
 `film_id` int(11) NOT NULL,
 `actor_id` int(11) NOT NULL,
 `remark` varchar(255) DEFAULT NULL,
 PRIMARY KEY (`id`),
 KEY `idx_film_actor_id` (`film_id`,`actor_id`)
 ) ENGINE=InnoDB DEFAULT CHARSET=utf8;

Let's take a look at the following sql:

select (select 1 from actor where id = 1) from (select * from film where id = 1) der;

1. It must be execution from Subsequent statements

2. In the execution of the inner layer select sentence

3. Execute the outer layer select sentence

We use it explain Look at the structure ：

From this we can deduce ：id The bigger it is, the earlier it is executed ,

Observe select_type：3. Is derived , That is, a temporary table    2 Is subquery 1. It is a primary key query

It can be seen that sql The order of execution is ： From the back to the front

 

type Column
This column represents the association type or access type , namely MySQL Decide how to look up rows in the table , Find the approximate range of data row records . From the best to the worst are ：system > const > eq_ref > ref > range > index > ALL
 

system What happened ： There is only one record in the query derived table

 

const What happened ：

The index has a primary key or a unique key

 

 eq_ref: The primary key and index are connected

 ref: The index is gone

 

 range: The scope of the index

 index: Scan the full index to get the results

 

ALL: Full table scan

 

key_len Column ：

key_len The calculation rules are as follows ：
char(n)： If you save Chinese characters, the length is 3n byte
varchar(n)： If you save Chinese characters, the length is 3n + 2 byte , Plus 2 Bytes are used to store string length , because
varchar Is a variable length string
value type
tinyint：1 byte
smallint：2 byte
int：4 byte
bigint：8 byte
Time type
date：3 byte
timestamp：4 byte
datetime：8 byte
If the field is allowed to be NULL, need 1 Is the byte record NULL
 

Why 140? 24*3+2+4+20*3+2=140

  Why 4？ Go, primary key , Other indexes do not take effect

10. Extra Column
Using where： Use where Statement to process the result , And the columns of the query are not covered by the index

 

 Using index： Use overlay index

Using index condition： The columns of the query are not completely covered by the index ,where In the condition is a leading range ;
 

Using filesort： Will use external sort instead of index sort , Sort data from memory when data is small , Otherwise, you need to finish sorting on disk

Select tables optimized away