JSON format execution plan (6) - MySQL execution plan (52)

We said that before extra, This mainly shows additional information , For example, if you don't fill in the form , Will be displayed no table, The index will show using index, Full table scanning or back table , Is displayed using where, If mysql Optimizer to internal query , It will also show the strategy of internal query selection , For example, internal connection temporary table to duplicate value query , For example, loose query , For example, the most principled method , Circular query .

Extra（5）—mysql Implementation plan （ Fifty-one )

If we were explain Add a format=JSON What will happen ？ At this time, we can get json data ,

mysql> EXPLAIN FORMAT=JSON SELECT * FROM s1 INNER JOIN s2 ON s1.key1 = s2.key2 WHERE s1.common_field = 'a'\G
*************************** 1. row ***************************
 
EXPLAIN: {
  "query_block": {
    "select_id": 1,     #  Whole select only one id It means 
    "cost_info": {
      "query_cost": "3197.16"   #  The whole inquiry cost is estimated to be 3197.16
    },
    "nested_loop": [    #  Nested loop join queries are used between several tables 
    
    #  The following is information about the tables involved in the nested loop join algorithm 
      {
        "table": {
          "table_name": "s1",   # s1 A table is a driver table 
          "access_type": "ALL",     #  The access method is ALL, It means using a full table scan to access 
          "possible_keys": [    #  Possible indexes 
            "idx_key1"
          ],
          "rows_examined_per_scan": 9688,   #  Check once s1 The table needs to be scanned roughly 9688 Bar record 
          "rows_produced_per_join": 968,    #  The driver table s1 The fan out is 968
          "filtered": "10.00",  # condition filtering Represents the percentage of 
          "cost_info": {
            "read_cost": "1840.84",     #  Explain later 
            "eval_cost": "193.76",      #  Explain later 
            "prefix_cost": "2034.60",   #  Single query s1 The total cost of the table 
            "data_read_per_join": "1M"  #  Amount of data read 
          },
          "used_columns": [     #  Execute the columns involved in the query 
            "id",
            "key1",
            "key2",
            "key3",
            "key_part1",
            "key_part2",
            "key_part3",
            "common_field"
          ],
          
          #  Yes s1 Table access conditions for single table queries 
          "attached_condition": "((`xiaohaizi`.`s1`.`common_field` = 'a') and (`xiaohaizi`.`s1`.`key1` is not null))"
        }
      },
      {
        "table": {
          "table_name": "s2",   # s2 A table is a driven table 
          "access_type": "ref",     #  The access method is ref, It means using index equivalent matching 
          "possible_keys": [    #  Possible indexes 
            "idx_key2"
          ],
          "key": "idx_key2",    #  Actual index used 
          "used_key_parts": [   #  Index column used 
            "key2"
          ],
          "key_length": "5",    # key_len
          "ref": [      #  And key2 Column the objects to be equivalently matched 
            "xiaohaizi.s1.key1"
          ],
          "rows_examined_per_scan": 1,  #  Check once s2 The table needs to be scanned roughly 1 Bar record 
          "rows_produced_per_join": 968,    #  Was the driver table s2 The fan out is 968（ Because there is no extra table to join , So this value doesn't work ）
          "filtered": "100.00",     # condition filtering Represents the percentage of 
          
          # s2 Table search criteria for queries using indexes 
          "index_condition": "(`xiaohaizi`.`s1`.`key1` = `xiaohaizi`.`s2`.`key2`)",
          "cost_info": {
            "read_cost": "968.80",      #  Explain later 
            "eval_cost": "193.76",      #  Explain later 
            "prefix_cost": "3197.16",   #  Word query s1, Multiple queries s2 Total cost 
            "data_read_per_join": "1M"  #  Amount of data read 
          },
          "used_columns": [     #  Execute the columns involved in the query 
            "id",
            "key1",
            "key2",
            "key3",
            "key_part1",
            "key_part2",
            "key_part3",
            "common_field"
          ]
        }
      }
    ]
  }
}
1 row in set, 2 warnings (0.00 sec)

Copy

Let's take a look at the driver table first S1 Of cost_info

Inside read_cost It's made up of two parts

1、 yes I/O cost

2、 testing rows * (1 - filter) Record cost of article (rows It is the output column of our previous execution plan ,rows stay json What the data shows is rows_examined_per_scan,filtered unchanged )

eval_cost： testing rows * filter Cost records .

prefix_cost It's a separate query s1 Cost of tables : read_cost + eval_cost.

data_read_per_join It represents the number of reads .

S2 Of cost_info Actually sum s1 almost , But because the driven table has the opportunity to be accessed many times , therefore read_cost and eval_cost It is accumulated by multiple visits , and prefix_cost yes s1 The cost and s2 The total cost of .

Extented Explain

in addition , When we run out explain after , And then you can use show warnings You can also see some extended information .

mysql> EXPLAIN SELECT s1.key1, s2.key1 FROM s1 LEFT JOIN s2 ON s1.key1 = s2.key1 WHERE s2.common_field IS NOT NULL;
+----+-------------+-------+------------+------+---------------+----------+---------+-------------------+------+----------+-------------+
| id | select_type | table | partitions | type | possible_keys | key      | key_len | ref               | rows | filtered | Extra       |
+----+-------------+-------+------------+------+---------------+----------+---------+-------------------+------+----------+-------------+
|  1 | SIMPLE      | s2    | NULL       | ALL  | idx_key1      | NULL     | NULL    | NULL              | 9954 |    90.00 | Using where |
|  1 | SIMPLE      | s1    | NULL       | ref  | idx_key1      | idx_key1 | 303     | xiaohaizi.s2.key1 |    1 |   100.00 | Using index |
+----+-------------+-------+------------+------+---------------+----------+---------+-------------------+------+----------+-------------+
2 rows in set, 1 warning (0.00 sec)
 
mysql> SHOW WARNINGS\G
*************************** 1. row ***************************
  Level: Note
   Code: 1003
Message: /* select#1 */ select `xiaohaizi`.`s1`.`key1` AS `key1`,`xiaohaizi`.`s2`.`key1` AS `key1` from `xiaohaizi`.`s1` join `xiaohaizi`.`s2` where ((`xiaohaizi`.`s1`.`key1` = `xiaohaizi`.`s2`.`key1`) and (`xiaohaizi`.`s2`.`common_field` is not null))
1 row in set (0.00 sec)

Copy

There are three fields ,level,code,message

code by 1003 when ,message Express sql Optimized statement , such as in Turn into half connection , Turn left connection into inner connection, etc .( Be careful , here message Just for reference , You can't run it directly )