SQLServer數據庫分頁查詢

一、創建表結構

CREATE TABLE [dbo].[a](
    [id] [int] NOT NULL,
    [name] [nvarchar](50) NULL,
    [age] [int] NULL)

二、添加測試數據

declare @i int
set @i=1
while(@i<10000)
begin
    insert into a select @i,left(newid(),7),12
    set @i += 1
end

三、分頁sql，下面例子是每頁10條，取第31-40條數據。

--方法一：Order by
Select Top (40-31+1)  * From a Where ID in (Select Top 40 ID From a Order by ID  ) Order by ID Desc

--方法二：not in/top
select top 10 * from a where id not in (select top 30 id from a order by id) order by id

--方法三：not exists 
select top 10 * from a where not exists (select 1 from (select top 30 id from a order by id)a1 where a1.id=a.id) order by id

--方法四:max/top
select top 10 * from a where id>(select max(id) from (select top 30 id from a order by id)a1) order by id

--方法五：row_number()
select top 10 * from (select row_number()over(order by id)rownumber,* from a)a1 where rownumber>30
select * from (select row_number()over(order by id)rownumber,* from a) a1 where rownumber>30 and rownumber<41
select * from (select row_number()over(order by id)rownumber,* from a)a1 where rownumber between 31 and 40

--方法六:row_number() 變體，不基於已有字段產生記錄序號，先按條件篩選以及排好序，再在結果集上給一常量列用於產生記錄序號
select *
from (select row_number()over(order by id)rownumber,* from (select top 40 * from a where 1=1 order by id)a
)b where rownumber>30

四、SQL語句效率測試

declare @begin_date datetime
declare @end_date datetime
select @begin_date = getdate()

<.....YOUR CODE.....>

select @end_date = getdate()
select datediff(ms,@begin_date,@end_date) as '毫秒'

1萬：基本感覺不到差異

10萬：

五、結論

1.max/top,ROW_NUMBER()都是比較不錯的分頁方法。相比ROW_NUMBER()只支持sql2005及以上版本，max/top有更好的可移植性，能同時適用於sql2000,access。

2.not exists感覺是要比not in效率高一點點。

3.ROW_NUMBER()的3種不同寫法效率看起來差不多。