[sql] efficient way to implement paging

Should I use LINQ's Skip() and Take() method for paging, or implement my own paging with a SQL query?

Which is most efficient? Why would I choose one over the other?

I'm using SQL Server 2008, ASP.NET MVC and LINQ.

The approach that I am giving is the fastest pagination that SQL server can achieve. I have tested this on 5 million records. This approach is far better than "OFFSET 10 ROWS FETCH NEXT 10 ROWS ONLY" provided by SQL Server.

-- The below given code computes the page numbers and the max row of previous page
-- Replace <<>> with the correct table data.
-- Eg. <<IdentityColumn of Table>> can be EmployeeId and <<Table>> will be dbo.Employees

DECLARE @PageNumber int=1; --1st/2nd/nth page. In stored proc take this as input param.
DECLARE @NoOfRecordsPerPage int=1000;

 DECLARE @PageDetails TABLE
       (
        <<IdentityColumn of Table>> int,
        rownum int,
        [PageNumber] int
       )           
       INSERT INTO @PageDetails values(0, 0, 0)
       ;WITH CTE AS
       (
       SELECT <<IdentityColumn of Table>>, ROW_NUMBER() OVER(ORDER BY <<IdentityColumn of Table>>) rownum FROM <<Table>>
       )
       Insert into @PageDetails 
       SELECT <<IdentityColumn of Table>>, CTE.rownum, ROW_NUMBER() OVER (ORDER BY rownum) as [PageNumber] FROM CTE WHERE CTE.rownum%@NoOfRecordsPerPage=0


--SELECT * FROM @PageDetails 

-- Actual pagination
SELECT TOP (@NoOfRecordsPerPage)
FROM <<Table>> AS <<Table>>
WHERE <<IdentityColumn of Table>> > (SELECT <<IdentityColumn of Table>> FROM 
@PageDetails WHERE PageNumber=@PageNumber)
ORDER BY <<Identity Column of Table>>

LinqToSql will automatically convert a .Skip(N1).Take(N2) into the TSQL syntax for you. In fact, every "query" you do in Linq, is actually just creating a SQL query for you in the background. To test this, just run SQL Profiler while your application is running.

The skip/take methodology has worked very well for me, and others from what I read.

Out of curiosity, what type of self-paging query do you have, that you believe is more efficient than Linq's skip/take?

you can further improve the performance, chech this

From CityEntities c
Inner Join dbo.MtCity t0 on c.CodCity = t0.CodCity
Where c.Row Between @p0 + 1 AND @p0 + @p1
Order By c.Row Asc

if you will use the from in this way it will give better result:

From   dbo.MtCity  t0
   Inner Join  CityEntities c on c.CodCity = t0.CodCity

reason: because you are using the where class on the CityEntities table which will eliminate many record before joining the MtCity, so 100% sure it will increase the performance many fold...

Anyway answer by rodrigoelp is really helpfull.

Thanks

Try using

FROM [TableX]
ORDER BY [FieldX]
OFFSET 500 ROWS
FETCH NEXT 100 ROWS ONLY

to get the rows from 501 to 600 in the SQL server, without loading them in memory. Note that this syntax has become available with SQL Server 2012 only

You can implement paging in this simple way by passing PageIndex

Declare @PageIndex INT = 1
Declare  @PageSize INT = 20

Select ROW_NUMBER() OVER ( ORDER BY Products.Name ASC )  AS RowNumber,
    Products.ID,
    Products.Name
into #Result 
From Products

SELECT @RecordCount = COUNT(*) FROM #Results 

SELECT * 
FROM #Results
WHERE RowNumber
BETWEEN
    (@PageIndex -1) * @PageSize + 1 
    AND
    (((@PageIndex -1) * @PageSize + 1) + @PageSize) - 1

In SQL Server 2008:

DECLARE @PAGE INTEGER = 2
DECLARE @TAKE INTEGER = 50

SELECT [t1].*
FROM (
    SELECT ROW_NUMBER() OVER (ORDER BY [t0].[COLUMNORDER] DESC) AS [ROW_NUMBER], [t0].*
    FROM [dbo].[TABLA] AS [t0]
    WHERE ([t0].[COLUMNS_CONDITIONS] = 1)
    ) AS [t1]
WHERE [t1].[ROW_NUMBER] BETWEEN ((@PAGE*@TAKE) - (@TAKE-1)) AND (@PAGE*@TAKE)
ORDER BY [t1].[ROW_NUMBER]

In t0 are all records In t1 are only those corresponding to that page

In 2008 we cant use Skip().Take()

The way is:

var MinPageRank = (PageNumber - 1) * NumInPage + 1
var MaxPageRank = PageNumber * NumInPage

var visit = Visita.FromSql($"SELECT * FROM (SELECT [RANK] = ROW_NUMBER() OVER (ORDER BY Hora DESC),* FROM Visita WHERE ) A WHERE A.[RANK] BETWEEN {MinPageRank} AND {MaxPageRank}").ToList();

We use a CTE wrapped in Dynamic SQL (because our application requires dynamic sorting of data server side) within a stored procedure. I can provide a basic example if you'd like.

I haven't had a chance to look at the T/SQL that LINQ produces. Can someone post a sample?

We don't use LINQ or straight access to the tables as we require the extra layer of security (granted the dynamic SQL breaks this somewhat).

Something like this should do the trick. You can add in parameterized values for parameters, etc.

exec sp_executesql 'WITH MyCTE AS (
    SELECT TOP (10) ROW_NUMBER () OVER ' + @SortingColumn + ' as RowID, Col1, Col2
    FROM MyTable
    WHERE Col4 = ''Something''
)
SELECT *
FROM MyCTE
WHERE RowID BETWEEN 10 and 20'

While LINQ-to-SQL will generate an OFFSET clause (possibly emulated using ROW_NUMBER() OVER() as others have mentioned), there is an entirely different, much faster way to perform paging in SQL. This is often called the "seek method" as described in this blog post here.

SELECT TOP 10 first_name, last_name, score
FROM players
WHERE (score < @previousScore)
   OR (score = @previousScore AND player_id < @previousPlayerId)
ORDER BY score DESC, player_id DESC

The @previousScore and @previousPlayerId values are the respective values of the last record from the previous page. This allows you to fetch the "next" page. If the ORDER BY direction is ASC, simply use > instead.

With the above method, you cannot immediately jump to page 4 without having first fetched the previous 40 records. But often, you do not want to jump that far anyway. Instead, you get a much faster query that might be able to fetch data in constant time, depending on your indexing. Plus, your pages remain "stable", no matter if the underlying data changes (e.g. on page 1, while you're on page 4).

This is the best way to implement paging when lazy loading more data in web applications, for instance.

Note, the "seek method" is also called keyset paging.