Question

我正在使用我的sql，我想删除表emaleikebusiness_2中的重复行，该表比较另一个名为per2的表中的数据我正在使用的查询

delete 
FROM 
     emaillikebusiness_2 
WHERE 
    emaillikebusiness_2.Email_id
  IN (
    SELECT 
    per2.BusinessEmail2
    FROM 
       per2 );

explain语句给出输出

'1', 'PRIMARY', 'emaillikebusiness_2', 'ALL', NULL, NULL, NULL, NULL, '30670', 'Using where'
'2', 'DEPENDENT SUBQUERY', 'per2', 'ALL', 'idx_email2person2', NULL, NULL, NULL, '24710', 'Range checked for each record (index map: 0x1)'

我还在两个表上创建了INDEX，并尝试在单个表上创建索引，但仍然需要很长时间才能执行查询。无论如何，我可以减少查询执行时间

Answer 1

MySQL倾向于优化WHERE ... IN (SELECT ...)。改为使用JOIN：

DELETE e.*
FROM emaillikebusiness_2 AS e
JOIN per2 AS p
ON e.Email_id = p.BusinessEmail2

我发现使用IN时，它会经常对第一个表进行全面扫描，在第二个表的索引中搜索Email_id，即使第一个表是比第二个大得多。但是通过JOIN，它可以在两个表的索引之间进行更有效的匹配。

Answer 2

试试这个：

DELETE emaillikebusiness_2.* FROM emaillikebusiness_2
 INNER JOIN per2 ON emaillikebusiness_2.Email_id = per2.BusinessEmail2

Because in most cases, Inner Join is significantly faster than IN clause.

查询需要很长时间才能执行

2 个答案: