April 30, 2014 at 4:37 am
Hi ,
Do you happen to know of a way to invalidate cached query plans?
I would rather target a specific query instead of invalidating all of them.
Also do you know of any sql server setting that will cause cached query plans to invalidate even though only one character in the queries has changed?
exec sp_executesql N'select
cast(5 as int) as DisplaySequence,
mt.Description + '' '' + ct.Description as Source,
c.FirstName + '' '' + c.LastName as Name,
cus.CustomerNumber Code,
c.companyname as "Company Name",
a.Address1,
a.Address2,
a.City,
sp.Abbreviation,
a.PostalCode,
cphone.PhoneNumber as Phone,
cfax.PhoneNumber as Fax,
ccell.PhoneNumber as Cell,
cemail.EmailAddress as Email,
wa.WebAddress as Web,
cl.Id as Id,
c.Id as ContactId
from
contactlink cl
left join ContactTypes ct on ct.id = cl.ContactTypeID
left join Contacts c on c.ID = cl.ContactID
left join Customers cus on cus.id = cl.CustomerID
left join MasterTypes mt on mt.id = ct.MasterTypeID
left join contactaddresslink cal on cal.ContactID = c.id
left join Addresses a on a.id = cal.AddressID
left join StatesProvinces sp on sp.id = a.StateProvinceID
left join ContactPhoneLink cpl_phone on cpl_phone.ContactID = c.id and cpl_phone.PhoneTypeID = 1
left join phones cphone on cphone.id=cpl_phone.PhoneID
left join ContactPhoneLink cpl_fax on cpl_fax.ContactID = c.id and cpl_fax.PhoneTypeID = 2
left join phones cFax on cFax.id=cpl_fax.PhoneID
left join ContactPhoneLink cpl_cell on cpl_cell.ContactID = c.id and cpl_cell.PhoneTypeID = 3
left join phones cCell on cCell.id=cpl_cell.PhoneID
left join ContactEmailLink cel on cel.ContactID = c.id
left join Emails cEmail on cEmail.id = cel.EmailID
left join ContactWebLink cwl on cwl.ContactID = c.id
left join WebAddresses wa on wa.id = cwl.WebAddressID
where cus.id = @customerId
',N'@CustomerId int',@CustomerId=1065
In this query we have seen (on some databases) simply changing ‘@CustomerId int',@CustomerId=1065’ too ‘@customerId int',@customerId=1065’ fixed the a speed problem….just changed the case on the Customer bind parameter. On other servers this has no effect. I’m thinking the server is using an old cached query plan, but don’t know for sure.
April 30, 2014 at 4:37 am
The query seems to have gone bad again on my machine with or without a capital ‘C’
I found a way to target and delete specific query plans, but it still is running slow from within .net.
I am trying to find out why it is slow sometimes and fast at other times, any help would be greatly appreciated.
Here is what I found
Find cached plans
SELECT plan_handle, st.text
FROM sys.dm_exec_cached_plans
CROSS APPLY sys.dm_exec_sql_text(plan_handle) AS st
WHERE text LIKE N'%wa.id = cwl.WebAddressID%'
Remove specific plan
DBCC FREEPROCCACHE (0x06001E00AC6A672F90961F6C0400000001000000000000000000000000000000000000000000000000000000
April 30, 2014 at 4:58 am
Please don't start multiple new threads for single problem.
No replies to this thread please, direct replies to http://www.sqlservercentral.com/Forums/Topic1566066-2799-1.aspx
Gail Shaw
Microsoft Certified Master: SQL Server, MVP, M.Sc (Comp Sci)
SQL In The Wild: Discussions on DB performance with occasional diversions into recoverability
April 30, 2014 at 5:02 am
It sounds like you're hitting bad parameter sniffing. But, without seeing the "good" and "bad" execution plans, it's hard to know for sure. You need to capture both plans and then look to the SELECT operator. There you'll find the compile time parameters. Look at the values there and compare those to your statistics to understand why you're getting a plan that is less than optimal. It has to do with the distribution of your data, or, possibly, the age of your statistics, or, again, possibly, how your statistics are generated (sampled vs. a full scan). I can't tell you much more than that based on the information presented.
To address the bad parameter sniffing, you need to determine if you need a single plan all the time, in which case using OPTIMIZE FOR some value would work. Or, if you need a specific plan every time the query runs, in which case using RECOMPILE is the way to go (depending of course on the compile time). Or, if you need a generic plan not based on specific values, in which case OPTIMIZE FOR UNKNOWN is the way to go. But again, I don't know what applies here since I can't see the plans, your statistics, or their sample rate.
"The credit belongs to the man who is actually in the arena, whose face is marred by dust and sweat and blood"
- Theodore Roosevelt
Author of:
SQL Server Execution Plans
SQL Server Query Performance Tuning
Viewing 4 posts - 1 through 3 (of 3 total)
You must be logged in to reply to this topic. Login to reply