SQL Server 2005 introduced the APPLY operator, which is very much like a join clause and which allows joining between two table expressions i.e. joining a left/outer table expression with a right/inner table expression. The difference between join and APPLY operator becomes evident when you have a table-valued expression on the right side and you want this table-valued expression to be evaluated for each row from the left table expression. In this tip I am going to demonstrate what APPLY operator is, how it differs from regular JOINs and what are few of its applications.
The APPLY operator allows you to join two table expressions; the right table expression is processed every time for each row from the left table expression. As you might have guessed, the left table expression is evaluated first and then right table expression is evaluated against each row of the left table expression for final result-set. The final result-set contains all the selected columns from the left table expression followed by all the columns of right table expression.
The APPLY operator comes in two variants, the CROSS APPLY and the OUTER APPLY. The CROSS APPLY operator returns only those rows from left table expression (in its final output) if it matches with right table expression. In other words, the right table expression returns rows for left table expression match only. Whereas the OUTER APPLY operator returns all the rows from left table expression irrespective of its match with the right table expression. For those rows for which there are no corresponding matches in right table expression, it contains NULL values in columns of right table expression. So you might now conclude, the CROSS APPLY is semantically equivalent to INNER JOIN (or to be more precise its like a CROSS JOIN with a correlated sub-query) with a implicit join condition of 1=1 whereas OUTER APPLY is semantically equivalent to LEFT OUTER JOIN.
You might be wondering if the same can be achieved with regular JOIN clause then why and when to use APPLY operator? Though the same can be achieved with normal JOIN, the need of APPLY arises if you have table-valued expression on right part and also in some cases use of APPLY operator boost the performance of your query. Let me explain you with help of some examples.
Script #1 creates a Department table to hold information about departments. Then it creates an Employee table which hold information about the employees. Please note, each employee belongs to a department, hence the Employee table has referential integrity with the Department table.
Script #1 - Creating some temporary objects to work on...
USE [tempdb] GO IF EXISTS (SELECT * FROM sys.objects WHERE OBJECT_ID = OBJECT_ID(N'[Employee]') AND type IN (N'U')) BEGIN DROP TABLE [Employee] END GO IF EXISTS (SELECT * FROM sys.objects WHERE OBJECT_ID = OBJECT_ID(N'[Department]') AND type IN (N'U')) BEGIN DROP TABLE [Department] END CREATE TABLE [Department]( [DepartmentID] [int] NOT NULL PRIMARY KEY, [Name] VARCHAR(250) NOT NULL, ) ON [PRIMARY] INSERT [Department] ([DepartmentID], [Name]) VALUES (1, N'Engineering') INSERT [Department] ([DepartmentID], [Name]) VALUES (2, N'Administration') INSERT [Department] ([DepartmentID], [Name]) VALUES (3, N'Sales') INSERT [Department] ([DepartmentID], [Name]) VALUES (4, N'Marketing') INSERT [Department] ([DepartmentID], [Name]) VALUES (5, N'Finance') GO CREATE TABLE [Employee]( [EmployeeID] [int] NOT NULL PRIMARY KEY, [FirstName] VARCHAR(250) NOT NULL, [LastName] VARCHAR(250) NOT NULL, [DepartmentID] [int] NOT NULL REFERENCES [Department](DepartmentID), ) ON [PRIMARY] GO INSERT [Employee] ([EmployeeID], [FirstName], [LastName], [DepartmentID]) VALUES (1, N'Orlando', N'Gee', 1 ) INSERT [Employee] ([EmployeeID], [FirstName], [LastName], [DepartmentID]) VALUES (2, N'Keith', N'Harris', 2 ) INSERT [Employee] ([EmployeeID], [FirstName], [LastName], [DepartmentID]) VALUES (3, N'Donna', N'Carreras', 3 ) INSERT [Employee] ([EmployeeID], [FirstName], [LastName], [DepartmentID]) VALUES (4, N'Janet', N'Gates', 3 )
First query in Script #2 selects data from Department table and uses CROSS APPLY to evaluate the Employee table for each record of the Department table. Second query simply joins the Department table with the Employee table and all the matching records are produced.
Script #2 - CROSS APPLY and INNER JOIN
SELECT * FROM Department D CROSS APPLY ( SELECT * FROM Employee E WHERE E.DepartmentID = D.DepartmentID ) A GO SELECT * FROM Department D INNER JOIN Employee E ON D.DepartmentID = E.DepartmentID GO
If you look at the results they produced, it is the exact same result-set; not only that even the execution plan for these queries are similar to each other and has equal query cost, as you can see in the image below. So what is the use of APPLY operator? How does it differ from a JOIN and how does it help in writing more efficient queries. I will discuss this later, but first let me show you an example of OUTER APPLY also.
The first query in Script #3 selects data from Department table and uses OUTER APPLY to evaluate the Employee table for each record of the Department table. For those rows for which there is not a match in Employee table, those rows contains NULL values as you can see in case of row 5 and 6. The second query simply uses a LEFT OUTER JOIN between the Department table and the Employee table. As expected the query returns all rows from Department table; even for those rows for which there is no match in the Employee table.
Script #3 - OUTER APPLY and LEFT OUTER JOIN
SELECT * FROM Department D OUTER APPLY ( SELECT * FROM Employee E WHERE E.DepartmentID = D.DepartmentID ) A GO SELECT * FROM Department D LEFT OUTER JOIN Employee E ON D.DepartmentID = E.DepartmentID GO
Even though the above two queries return the same information, the execution plan is a bit different. Although cost wise there is not much difference, the query with the OUTER APPLY uses a Compute Scalar operator (which has an estimated operator cost of 0.0000103 or almost 0% of total query cost) before Nested Loops operator to evaluate and produce the columns of Employee table.
Now comes the time to see where the APPLY operator is really required. In Script #4, I am creating a table-valued function which accepts DepartmentID as its parameter and returns all the employees who belong to this department. The next query selects data from Department table and uses CROSS APPLY to join with the function we created. It passes the DepartmentID for each row from the outer table expression (in our case Department table) and evaluates the function for each row similar to a correlated subquery. The next query uses the OUTER APPLY in place of CROSS APPLY and hence unlike CROSS APPLY which returned only correlated data, the OUTER APPLY returns non-correlated data as well, placing NULLs into the missing columns.
Script #4 - APPLY with table-valued function
IF EXISTS (SELECT * FROM sys.objects WHERE OBJECT_ID = OBJECT_ID(N'[fn_GetAllEmployeeOfADepartment]') AND type IN (N'IF')) BEGIN DROP FUNCTION dbo.fn_GetAllEmployeeOfADepartment END GO CREATE FUNCTION dbo.fn_GetAllEmployeeOfADepartment(@DeptID AS INT) RETURNS TABLE AS RETURN ( SELECT * FROM Employee E WHERE E.DepartmentID = @DeptID ) GO SELECT * FROM Department D CROSS APPLY dbo.fn_GetAllEmployeeOfADepartment(D.DepartmentID) GO SELECT * FROM Department D OUTER APPLY dbo.fn_GetAllEmployeeOfADepartment(D.DepartmentID) GO
So now if you are wondering, can we use a simple join in place of the above queries? Then the answer is NO, if you replace CROSS/OUTER APPLY in the above queries with INNER JOIN/LEFT OUTER JOIN, specify ON clause (something as 1=1) and run the query, you will get "The multi-part identifier "D.DepartmentID" could not be bound." error. This is because with JOINs the execution context of outer query is different from the execution context of the function (or a derived table), and you can not bind a value/variable from the outer query to the function as a parameter. Hence the APPLY operator is required for such queries.
So in summary the APPLY operator is required when you have to use table-valued function in the query, but it can also be used with an inline SELECT statements.
Now let me show you another query with a Dynamic Management Function (DMF). Script #5 returns all the currently executing user queries except for the queries being executed by the current session. As you can see the script below, the sys.dm_exec_requests dynamic management view is being CROSS APPLY'ed with the sys.dm_exec_sql_text dynamic management function which accepts a "plan handle" for the query and the same "plan handle" is being passed from the left/outer expression to the function to work and to return the data.
Script #5 - APPLY with Dynamic Management Function (DMF)
USE master GO SELECT DB_NAME(database_id) AS [Database], [text] AS [Query] FROM sys.dm_exec_requests r CROSS APPLY sys.dm_exec_sql_text(r.plan_handle) st WHERE session_Id > 50 -- Consider spids for users only, no system spids. AND session_Id NOT IN (@@SPID) -- Don't include request from current spid.
Please note the [text] column in the above query returns the all queries submitted in a batch; if you want to see only active (currently executing) query you can use statement_start_offset and statement_end_offset columns to trim the active part of the query. Tim Ford has provided a very good explanation of usage of these columns in his How to isolate the current running commands in SQL Server tip.
As I told you before there are certain scenarios where a query with APPLY operator performs better than a query with regular joins but I am not going to delve into much details rather here are some articles which discuss this topic in greater details.
Please note, APPLY operator is not an ANSI operator but rather an extension of SQL Server T-SQL (available in SQL Server 2005 and above), so if you plan to port your database to some other DBMS take this into your considerations.
I have one scalar function.. So i need to update the column row by row by using the output of the function .I tried to use the Cross apply by using this function. its not working for me.It says invalid object name. even though object is there. Can't we use the scalar function in cross apply ? Any help much appriciated..
*** NOTE *** - If you want to include code from SQL Server Management Studio (SSMS) in your post, please copy the code from SSMS and paste the code into a text editor like NotePad before copying the code below to remove the SSMS formatting
Hmm, seems that the writer nor the readers have understood the topic. The NO-answer to the critical question is wrong. The answer should be YES which takes the point from the whole article. A JOIN would do the same thing.
Tuesday, July 22, 2014 - 3:51:39 AM - Satyendra Verma
I was accidently going through your article when i was searching solution for joining table and functions. I was wondering in early 2000 how nice it would be if we are able to join table and function and i am glad to see the same now. i was out of touch with SQL server for few years. When i was going through your article, i was thinking CROSS APPLY & OUTER APPLY are equal to JOIN & LEFT OUTER JOIN... i could able to see the same comment in your article... also i was thinking of the special purpose which cannot be acheived in JOINs which can be acheived in APPLY and i was able to get clear idea about the same through your excellent examples.
quote... "So now if you are wondering, can we use a simple join in place of the above queries? Then the answer is NO, if you replace CROSS/OUTER APPLY in the above queries with INNER JOIN/LEFT OUTER JOIN, specify ON clause (something as 1=1) and run the query, you will get "The multi-part identifier "D.DepartmentID" could not be bound."
I have to admit, I dont understand - it appears to me that the result set is exactly the same as the one for inner and that the answer to the question you ask is YES, not NO.
Nice article, I have been forced to use outer apply today so thought of reading an article related to the Apply operator... you have witten the article in simple words anybody can understand..keep posting...
MVP in ASP.NET/IIS
Thursday, July 12, 2012 - 2:30:45 AM - Brian Kirton
What this does is for each user in sysusers, returns all the columns and also all the columns for the latest object they have created. Try to write this without the outer apply and you will see why it is useful.
I'm confused. I still don't know why I would use APPLY. Your examples show a more complicated way to create inner and outer joins, but I fail to see the benefit. Please show an example that demonstrates beneifits not found in simple joins.
Thanks for the overview, but I'm still unsure about the need for cross apply. In your example you state that simple joins would not accomplish your cross apply results:
SELECT * FROM Department D CROSS APPLY dbo.fn_GetAllEmployeeOfADepartment(D.DepartmentID)
But the following seems to do just that in a simpler and more readable format (because functionality is not black-boxed in a function):
select * from department d inner join employee e on d.departmentid=e.departmentid
If the performance is the same and the code is more readable, why was CROSS APPLY ever created and why use it? After reading several articles on this, I suspect that CROSS APPLY was created because table-valued functions are extremely limited without CROSS/OUTER APPLY. This is readily apparent when you look at dynamic management functions (DMF) which are two-dimensional and hobbled without CROSS APPLY functionality.
So I guess this begs a larger question: are table-valued functions necessary? I thought they were a great idea, but I've yet to actually use one in a production environment. I've never needed them.
Wednesday, July 13, 2011 - 10:31:56 AM - Alexander