为什么这个索引搜索会导致比索引扫描更多的读取

正如我所读到的，大多数时候索引搜索比索引扫描更受欢迎，我正在尝试一些东西。

我有一个查询在使用索引扫描时进行 993 次读取（使用 SQL Profiler 检查）。使用索引查找时，它需要 44.347 次读取。感觉有些不对劲，或者我不明白。

这是索引扫描的查询：

select      t5.Id as t5Id
from        table1 t1
left join   table2 t2 on t2.Table1Id = t1.Id
left join   table3 t3 on t3.Table2Id = t2.Id
left join   table4 t4 on t4.Table3Id = t3.Id
left join   table5 t5 on t5.Table4Id = t4.Id

这是索引查找的查询：

select      t5.Id as t5Id
from        table1 t1
left join   table2 t2 on t2.Table1Id = t1.Id
left join   table3 t3 on t3.Table2Id = t2.Id
left join   table4 t4 on t4.Table3Id = t3.Id
left join   table5 t5 WITH (FORCESEEK) on t5.Table4Id = t4.Id

这些表格简单明了。最后，我用一些虚拟数据填充它们，因此可以轻松复制。

CREATE TABLE [dbo].[table1](
    [Id] [bigint] IDENTITY(1,1) NOT NULL,
    [Name] [nvarchar](max) NOT NULL,
CONSTRAINT [PK_table1] 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] TEXTIMAGE_ON [PRIMARY]
GO




CREATE TABLE [dbo].[table2](
    [Id] [bigint] IDENTITY(1,1) NOT NULL,
    [Table1Id] [bigint] NOT NULL,
CONSTRAINT [PK_table2] 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]

ALTER TABLE     [dbo].[table2] WITH CHECK
ADD CONSTRAINT  [FK_table2_table1Id] FOREIGN KEY([table1Id])
REFERENCES      [dbo].[table1] ([Id])
GO

ALTER TABLE         [dbo].[table2]
CHECK CONSTRAINT    [FK_table2_table1Id]
GO


CREATE NONCLUSTERED INDEX [IdxTable2_FKTable1Id] ON [dbo].[table2]
(
    [Table1Id] ASC
)
INCLUDE (   [Id]) WITH (PAD_INDEX = OFF, STATISTICS_NORECOMPUTE = OFF, SORT_IN_TEMPDB = OFF, DROP_EXISTING = OFF, ONLINE = OFF, ALLOW_ROW_LOCKS = ON, ALLOW_PAGE_LOCKS = ON) ON [PRIMARY]
GO






CREATE TABLE [dbo].[table3](
    [Id] [bigint] IDENTITY(1,1) NOT NULL,
    [Table2Id] [bigint] NOT NULL,
CONSTRAINT [PK_table3] 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]

ALTER TABLE     [dbo].[table3] WITH CHECK
ADD CONSTRAINT  [FK_table3_table2Id] FOREIGN KEY([table2Id])
REFERENCES      [dbo].[table2] ([Id])
GO

ALTER TABLE         [dbo].[table3]
CHECK CONSTRAINT    [FK_table3_table2Id]
GO


CREATE NONCLUSTERED INDEX [IdxTable3_FKTable2Id] ON [dbo].[table3]
(
    [Table2Id] ASC
)
INCLUDE (   [Id]) WITH (PAD_INDEX = OFF, STATISTICS_NORECOMPUTE = OFF, SORT_IN_TEMPDB = OFF, DROP_EXISTING = OFF, ONLINE = OFF, ALLOW_ROW_LOCKS = ON, ALLOW_PAGE_LOCKS = ON) ON [PRIMARY]
GO







CREATE TABLE [dbo].[table4](
    [Id] [bigint] IDENTITY(1,1) NOT NULL,
    [Table3Id] [bigint] NOT NULL,
CONSTRAINT [PK_table4] 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]

ALTER TABLE     [dbo].[table4] WITH CHECK
ADD CONSTRAINT  [FK_table4_table3Id] FOREIGN KEY([table3Id])
REFERENCES      [dbo].[table4] ([Id])
GO

ALTER TABLE         [dbo].[table4]
CHECK CONSTRAINT    [FK_table4_table3Id]
GO

CREATE NONCLUSTERED INDEX [IdxTable4_FKTable3Id] ON [dbo].[table4]
(
    [Table3Id] ASC
)
INCLUDE (   [Id]) WITH (PAD_INDEX = OFF, STATISTICS_NORECOMPUTE = OFF, SORT_IN_TEMPDB = OFF, DROP_EXISTING = OFF, ONLINE = OFF, ALLOW_ROW_LOCKS = ON, ALLOW_PAGE_LOCKS = ON) ON [PRIMARY]
GO





CREATE TABLE [dbo].[table5](
    [Id] [bigint] IDENTITY(1,1) NOT NULL,
    [Table4Id] [bigint] NOT NULL,
    [Description] [nvarchar](2000) NOT NULL,
CONSTRAINT [PK_table5] 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]

ALTER TABLE     [dbo].[table5] WITH CHECK
ADD CONSTRAINT  [FK_table5_table4Id] FOREIGN KEY([table4Id])
REFERENCES      [dbo].[table5] ([Id])
GO

ALTER TABLE         [dbo].[table5]
CHECK CONSTRAINT    [FK_table5_table4Id]
GO

CREATE NONCLUSTERED INDEX [IdxTable5_FKTable4Id] ON [dbo].[table5]
(
    [Table4Id] ASC
)
INCLUDE (   [Id], [Description]) WITH (PAD_INDEX = OFF, STATISTICS_NORECOMPUTE = OFF, SORT_IN_TEMPDB = OFF, DROP_EXISTING = OFF, ONLINE = OFF, ALLOW_ROW_LOCKS = ON, ALLOW_PAGE_LOCKS = ON) ON [PRIMARY]
GO


set nocount on

DECLARE @i INT = 0;
DECLARE @j INT = 0;
DECLARE @k INT = 0;
DECLARE @l INT = 10;
DECLARE @m INT = 0;

declare @table1Id bigint
declare @table2Id bigint
declare @table3Id bigint
declare @table4Id bigint


begin tran
WHILE @i < 10
BEGIN
    INSERT INTO [dbo].[table1] ([Name]) VALUES (cast(@i as nvarchar(10)))
    SELECT @table1Id = SCOPE_IDENTITY()

    WHILE @j < 10
    BEGIN
        INSERT INTO [dbo].[table2] ([Table1Id]) VALUES (@table1Id)
        SELECT @table2Id = SCOPE_IDENTITY()

        WHILE @k < 10
        BEGIN
            INSERT INTO [dbo].[table3] ([Table2Id]) VALUES (@table2Id)
            SELECT @table3Id = SCOPE_IDENTITY()

            WHILE @l > 0
            BEGIN
                INSERT INTO [dbo].[table4] ([Table3Id]) VALUES (@table3Id)
                SELECT @table4Id = SCOPE_IDENTITY()

                WHILE @m < 10
                BEGIN
                    INSERT INTO [dbo].[table5] ([Table4Id], [Description]) VALUES (@table4Id, 'Not so long description')

                    SET @m = @m + 1;
                END;

                SET @m = 0;
                SET @l = @l - 1;
            END;

            SET @l = 10;
            SET @k = @k + 1;
        END;

        SET @k = 0;
        SET @j = @j + 1;
    END;

    SET @j = 0;
    SET @i = @i + 1;
END;
commit