使用group by优化子查询的sql

时间:2014-03-25 07:40:34

标签: mysql sql

有人可以帮我优化这个查询吗?我有一张包含大量数据的表格。有人可以帮我优化而不使用子查询吗?

SELECT user_id, scheduled_on_date
  FROM
    (SELECT user_id, scheduled_on_date
     FROM `calls`
     ORDER BY scheduled_on_date DESC) AS cinner
  GROUP BY user_id

预期产出为: 我需要的是用户的唯一行以及上次调度的调用日期。

2 个答案:

答案 0 :(得分:1)

您可以按如下方式重写查询

SELECT c.user_id, c.scheduled_on_date ,other_fields_max_per_group
FROM `calls` c
JOIN (SELECT user_id, MAX(scheduled_on_date) scheduled_on_date
     FROM `calls`
      GROUP BY user_id) AS cc
ON(c.user_id =cc.user_id AND c.scheduled_on_date =cc.scheduled_on_date)

添加化合物索引

ALTER TABLE calls ADD INDEX `test` (user_id ,scheduled_on_date )

如果您只想要最大的日期和用户ID,则只能使用

SELECT user_id, MAX(scheduled_on_date) scheduled_on_date
FROM `calls`
GROUP BY user_id

答案 1 :(得分:0)

select c1.*
from calls c1 left join call c2 on c1.user_id = c2.user_id 
     and c1.scheduled_on_date < c2.scheduled_on_date
where c2.user_id is null;

如果您想进一步优化

  • 确保您拥有索引(user_id,scheduled_on_date)
  • 仅获取所需的列