Question

我有很多像这样的问题......

SELECT colA, colB, colC, colD FROM table1 
  LEFT JOIN (
    SELECT colD FROM table2 WHERE colE = 3225 
  ) AS subquery
ON colD = colA ;

在我的服务器的“log-queries-not-using-indexes”日志中列出了哪些。

EXPLAIN看起来像这样......

id | select_type | table      | type  | possible_keys | key     | key_len | ref  | rows | Extra
---+-------------+------------+-------+---------------+---------+---------+------+------+-------------
 1 | PRIMARY     | table1     | range | PRIMARY       | PRIMARY | 3       | NULL |   58 | Using where
 1 | PRIMARY     | <derived2> | ALL   |          NULL |    NULL |    NULL | NULL |    1 | 
 2 | DERIVED     | table2     | const | PRIMARY       | PRIMARY | 3       |      |    1 |

我有什么办法可以重构Query，还是我应该添加任何新的索引来改进这个？

仅供参考表格的结构如下......

CREATE TABLE `table1` (
 `colA` MEDIUMINT( 9 ) NOT NULL ,
 `colB` VARCHAR( 100 ) DEFAULT NULL ,
 `colC` VARCHAR( 6 ) DEFAULT NULL ,
 some more columns removed to simplify things...
 PRIMARY KEY (  `colA` ) ,
 KEY  `colB` (  `colB` ) ,
 KEY  `colC` (  `colC` ) 
) ENGINE = MYISAM DEFAULT CHARSET = latin1 ;


CREATE  TABLE `table2` (
 `colE` mediumint( 9  )  NOT  NULL  auto_increment ,
 `colD` mediumint( 9  ) default  '0',
 some more columns removed to simplify things...
 PRIMARY  KEY (  `colE`  ) ,
 KEY  `colD` (  `colD`  )  
) ENGINE  =  MyISAM  DEFAULT CHARSET  = latin1 ;

提前致谢，

此致

菲尔

Answer 1

我会尝试以下方法：

SELECT ColA, ColB, ColC, ColD
FROM
    table1
    LEFT JOIN table2
        ON ColE = 3225
        AND ColD = ColA

另外，尝试在ColE上创建涵盖ColD和table2的索引：

CREATE INDEX `IDX_table2` on `table2` (`colE` ASC, `colD` ASC);

使用左连接和子查询优化MySQL查询

1 个答案: