使用ROW_NUMBER()为每个UserId按递减日期分配唯一的排名，然后为每个UserId过滤到第一行(即ROW_NUMBER = 1)。

SELECT UserId, Value, Date
FROM (SELECT UserId, Value, Date,
        ROW_NUMBER() OVER (PARTITION BY UserId ORDER BY Date DESC) rn
      FROM users) u
WHERE rn = 1;

我想是这样的。(请原谅我的语法错误;在这一点上，我习惯使用HQL !)

编辑:也误解了问题!修正了查询…

SELECT UserId, Value
FROM Users AS user
WHERE Date = (
    SELECT MAX(Date)
    FROM Users AS maxtest
    WHERE maxtest.UserId = user.UserId
)

难道一个qualified子句不是既简单又最好吗?

select userid, my_date, ...
from users
qualify rank() over (partition by userid order by my_date desc) = 1

对于上下文，在Teradata这里一个像样的大小测试运行在17秒与这个合格版本和在23秒与“内联视图”/Aldridge解决方案#1。

MySQL没有分区KEEP, DENSE_RANK概念的解决方案。

select userid,
       my_date,
       ...
from
(
select @sno:= case when @pid<>userid then 0
                    else @sno+1
    end as serialnumber, 
    @pid:=userid,
       my_Date,
       ...
from   users order by userid, my_date
) a
where a.serialnumber=0

参考:http://benincampus.blogspot.com/2013/08/select-rows-which-have-maxmin-value-in.html

SELECT a.* 
FROM user a INNER JOIN (SELECT userid,Max(date) AS date12 FROM user1 GROUP BY userid) b  
ON a.date=b.date12 AND a.userid=b.userid ORDER BY a.userid;

我不知道你的列的确切名称，但它应该是这样的:

SELECT userid, value
FROM users u1
WHERE date = (
    SELECT MAX(date)
    FROM users u2
    WHERE u1.userid = u2.userid
)