到目前为止我发现了什么：
Microsoft’s Definition“使用一组特定的输入值调用确定性函数时,它们总是返回相同的结果,并给出相同的数据库状态.”

所以在Set理论表中
雇员

Employee            Salary
Sue Right            1.00
Robin Page           1.00
Phil Factor          1.00

和
雇员2

Employee            Salary
Phil Factor          1.00
Sue Right            1.00
Robin Page           1.00

是相同的.但排名函数返回不同的值：

CREATE TABLE [dbo].[Employees](
    --[ID] [int] IDENTITY(1,1) NOT NULL,[Employee] [varchar](150) NOT NULL,[Salary] [smallmoney] NULL,) ON [PRIMARY]

GO
CREATE TABLE [dbo].[Employees2](
    --[ID] [int] IDENTITY(1,) ON [PRIMARY]

INSERT INTO [dbo].[Employees]
([Employee],[Salary])
VALUES
('Sue Right',1),('Robin Page',('Phil Factor',1 )
GO
INSERT INTO [dbo].[Employees2]
([Employee],[Salary])
VALUES
('Phil Factor',1 ),('Sue Right',1)
GO
SELECT RANK() OVER ( ORDER BY Salary) AS [Rank],DENSE_RANK() OVER (ORDER BY Salary ) AS [Dense_rank],[Employee]
FROM
dbo.Employees

SELECT RANK() OVER ( ORDER BY Salary) AS [Rank],[Employee]
FROM
dbo.Employees2

SELECT NTILE(3) OVER ( ORDER BY SALARY ),[Employee]
FROM
dbo.Employees

SELECT NTILE(3) OVER ( ORDER BY SALARY ),[Employee]
FROM
dbo.Employees2