小编Apr*_*llo的帖子

我有下表，其中有 750 万条记录：

CREATE TABLE [dbo].[TestTable](
    [Id] [int] IDENTITY(1,1) NOT NULL,
    [TestCol] [nvarchar](50) NOT NULL,
    [TestCol2] [nvarchar](50) NOT NULL,
    [TestCol3] [nvarchar](50) NOT NULL,
    [Anonymised] [tinyint] NOT NULL,
    [Date] [datetime] NOT NULL,
CONSTRAINT [PK_TestTable] PRIMARY KEY CLUSTERED 
(
    [Id] ASC
)WITH (PAD_INDEX = OFF, STATISTICS_NORECOMPUTE = OFF, IGNORE_DUP_KEY = OFF, 
ALLOW_ROW_LOCKS = ON, ALLOW_PAGE_LOCKS = ON) ON [PRIMARY]
) ON [PRIMARY]

我注意到当日期字段上有非聚集索引时：

CREATE NONCLUSTERED INDEX IX_TestTable_Date ON [dbo].[TestTable] ([Date])

- 我运行以下查询：

UPDATE TestTable 
SET TestCol='*GDPR*', TestCol2='*GDPR*', TestCol3='*GDPR*', Anonymised=1
WHERE [Date] <= …

我正在一个有 750 万条记录的数据库上运行以下查询：

UPDATE TestTable
SET TestCol='1234', TestCol2='1234', TestCol3='1234', Anonymised=1
WHERE [Date] >= '25 June 2016' 
AND [Date] <= '25 August 2016'

出于某种原因，执行计划包括对主键的排序操作。

鉴于该列上已经有聚集索引，我不希望它必须对主键进行排序。

任何人都可以解释为什么会发生这种情况以及我可以做些什么来避免排序操作？

编辑：

表创建脚本

CREATE TABLE [dbo].[TestTable](
    [Id] [int] IDENTITY(1,1) NOT NULL,
    [TestCol] [nvarchar](50) NOT NULL,
    [TestCol2] [nvarchar](50) NOT NULL,
    [TestCol3] [nvarchar](50) NOT NULL,
    [Anonymised] [tinyint] NOT NULL,
    [Date] [datetime] NOT NULL,
 CONSTRAINT [PK_TestTable] PRIMARY KEY CLUSTERED 
(
    [Id] ASC
)WITH (PAD_INDEX = OFF, STATISTICS_NORECOMPUTE = OFF, IGNORE_DUP_KEY = OFF, 
ALLOW_ROW_LOCKS = ON, ALLOW_PAGE_LOCKS = ON) ON …