Different execution plan between SQL and SP

SQLServerCentral is supported by Red Gate Software Ltd.

Popular Topics

Home

Members

Calendar

Who's On

Home » SQL Server 2008 » SQL Server 2008 - General » Different execution plan between SQL and SP

12 posts, Page 1 of 2

1 2 »»

Different execution plan between SQL and SP

Rate Topic

Display Mode

Topic Options

Author

Message

SGT_squeequal

Posted Tuesday, November 27, 2012 12:59 PM

SSC Veteran

Group: General Forum Members
Last Login: Saturday, March 30, 2013 9:39 AM
Points: 261, Visits: 966

Hello, wonder if someone can help, i am looking at a performance issue with a stored procedure (SP). The SP its self is very simple only a few table joins and a single parameter ( integer data type).
When I execute as a stored procedure I get a totally different execution plan to when I execute as a select statement. Both queries are executed using the same method through SSMS QA
Below is an extract from statistics profiles, set one is from the SQL query, the corresponding execution plan to this query uses clustered index seeks whereas the second set of data executing as SP, you can see there are a lot of index scans. However, if i execute the SP as Exec SP_NAME PARAMETER with Recompile the query executes using the better execution plan.
AS SQL Query
Parent PhysicalOp LogicalOp
0 NULL NULL
1 Nested Loops Inner Join
2 Compute Scalar Compute Scalar
3 Nested Loops Inner Join
4 Nested Loops Inner Join
5 Nested Loops Inner Join
6 Compute Scalar Compute Scalar
7 Nested Loops Inner Join
8 Nested Loops Inner Join
9 Clustered Index Seek Clustered Index Seek
9 Filter Filter
11 Clustered Index Seek Clustered Index Seek
8 Clustered Index Seek Clustered Index Seek
6 Clustered Index Seek Clustered Index Seek
5 Clustered Index Seek Clustered Index Seek
4 Clustered Index Seek Clustered Index Seek
2 Clustered Index Seek Clustered Index Seek

AS SP
Parent PhysicalOp LogicalOp
0 NULL NULL
1 Sort Sort
2 Parallelism Gather Streams
3 Filter Filter
4 Hash Match Right Outer Join
5 Parallelism Repartition Streams
6 Index Scan Index Scan
5 Hash Match Inner Join
8 Bitmap Bitmap Create
9 Parallelism Repartition Streams
10 Hash Match Inner Join
11 Parallelism Repartition Streams
12 Clustered Index Scan Clustered Index Scan
11 Hash Match Inner Join
14 Parallelism Repartition Streams
15 Clustered Index Scan Clustered Index Scan
14 Parallelism Repartition Streams
17 Hash Match Inner Join
18 Compute Scalar Compute Scalar
19 Parallelism Repartition Streams
20 Clustered Index Scan Clustered Index Scan
18 Parallelism Repartition Streams
25 Index Scan Index Scan
8 Hash Match Inner Join
30 Bitmap Bitmap Create
31 Parallelism Repartition Streams
32 Index Scan Index Scan
30 Compute Scalar Compute Scalar
36 Parallelism Repartition Streams
37 Clustered Index Scan Clustered Index Scan

I have updated the statistics and added in a second variable inside the SP advice i have used in passed for parameter sniffing. Anyone come across this before and can anyone help shed some light on what is happening.

Cheers

*************************************************************

The first is always the hardest w00t

Post #1389388

GilaMonster

Posted Tuesday, November 27, 2012 1:18 PM

SSC-Dedicated

Group: General Forum Members
Last Login: Today @ 12:47 PM
Points: 37,730, Visits: 29,992

Can you post the graphical plans please?

Gail Shaw
Microsoft Certified Master: SQL Server 2008, MVP
SQL In The Wild: Discussions on DB performance with occasional diversions into recoverability

We walk in the dark places no others will enter
We stand on the bridge and no one may pass

Post #1389393

SGT_squeequal

Posted Tuesday, November 27, 2012 1:40 PM

SSC Veteran

Group: General Forum Members
Last Login: Saturday, March 30, 2013 9:39 AM
Points: 261, Visits: 966

HI Gail,

here you go thank you for you help, am tearing my hair out to try understand this :)

*************************************************************

The first is always the hardest w00t

Post Attachments

SP_PLAN.sqlplan (14 views, 89.87 KB)

SELECT_PLAN.sqlplan (9 views, 44.06 KB)

Post #1389402

SQLSACT

Posted Tuesday, November 27, 2012 11:46 PM

Ten Centuries

Group: General Forum Members
Last Login: Today @ 7:48 AM
Points: 1,200, Visits: 2,126

Im no Guru but I think this could be Bad Parameter Sniffing

http://sqlinthewild.co.za/index.php/2007/11/27/parameter-sniffing/

Post #1389520

Dave Ballantyne

Posted Wednesday, November 28, 2012 12:24 AM

SSCommitted

Group: General Forum Members
Last Login: Friday, May 10, 2013 4:07 PM
Points: 1,943, Visits: 8,227

Hmm,

There is a slight difference in the code , the 'fast' version includes the extra clause @pJobId > 0
The optimizer is probably promoting this to the JobWorkJob.JobOID which give a differenting plan.

FROM
              JobStay
               JOIN UDStay  ON JobStay.JobOID = UDStay.StayOID
               JOIN Job  ON JobStay.JobOID = Job.JobOID
               JOIN Item  ON JobStay.ItemOID = Item.ItemOID
	LEFT JOIN JobWorkJob ON JobWorkJob.ForJobOID = JobStay.JobOID
	LEFT JOIN JOB WJ ON WJ.JobOID = JobWorkJob.JobOID
	JOIN Commodity ON Commodity.CommodityOID = Item.CommodityOID
	JOIN UDCommodity ON UDCommodity.CommodityOID = Commodity.CommodityOID




WHERE
	@pJobOID > 0
	AND
	JobWorkJob.JobOID =  @pJobOID

However.....

Is this really what you want to do ?
You are left joining to JobWorkJob and then filtering the result!
Should that be an inner join ?
Or an outer join ON ID and ID and JobOID = @pJobOID

Clear Sky SQL
My Blog
Kent user group

Post #1389535

SGT_squeequal

Posted Wednesday, November 28, 2012 12:36 AM

SSC Veteran

Group: General Forum Members
Last Login: Saturday, March 30, 2013 9:39 AM
Points: 261, Visits: 966

HI,

thanks for your reply but i took that extra parameter @joboid>0 out of the SP as its was a usless parameter not sure what the reason behind the person who wrote it putting it in.

*************************************************************

The first is always the hardest w00t

Post #1389543

Dave Ballantyne

Posted Wednesday, November 28, 2012 1:19 AM

SSCommitted

Group: General Forum Members
Last Login: Friday, May 10, 2013 4:07 PM
Points: 1,943, Visits: 8,227

Maybe useless to you :) but the optimizer is chosing to do something different because of it.

What about the outer join ? Is that what is intended ?

Clear Sky SQL
My Blog
Kent user group

Post #1389556

SGT_squeequal

Posted Wednesday, November 28, 2012 2:24 AM

SSC Veteran

Group: General Forum Members
Last Login: Saturday, March 30, 2013 9:39 AM
Points: 261, Visits: 966

I have changed the joins now to inner joins as i realize the where clause negates on of the left joins and the other can also be inner join, not sure why they wrote this like it does i have a few more things to try with the parameter sniffing will give them a go :)

*************************************************************

The first is always the hardest w00t

Post #1389587

Dave Ballantyne

Posted Wednesday, November 28, 2012 3:09 AM

SSCommitted

Group: General Forum Members
Last Login: Friday, May 10, 2013 4:07 PM
Points: 1,943, Visits: 8,227

So after you change it to an inner join , are you still getting different plans ?

Clear Sky SQL
My Blog
Kent user group

Post #1389609

SGT_squeequal

Posted Wednesday, November 28, 2012 3:23 AM

SSC Veteran

Group: General Forum Members
Last Login: Saturday, March 30, 2013 9:39 AM
Points: 261, Visits: 966

Yup i still have the same issue

if i run the SP like

Exec SP_NAME parameter_value with recompile

the SP run fine and uses the good execution plan

if i run it with

Exec SP_NAME parameter_value

it uses the poor execution plan

If i add the option to recompile within the SP and execute

Exec SP_NAME parameter_value

then i still get the poor execution plan w00t

Done a few changes to the SP to use local variables to try stop any parameter sniffing, no joy, tired the option (optimize for Unknown ) and still no joy Sad

*************************************************************

The first is always the hardest w00t

Post #1389613

« Prev Topic | Next Topic »

12 posts, Page 1 of 2

1 2 »»

Permissions