Thursday 21 September 2017

Quickly find a cached execution plan of an Ad-Hoc query

It is clear that sometimes we just need to find a cached execution plan as quickly as possible so that it can be analysed and then use it to optimise the code in question. This work might be harder if the workload of a database server is based on ad-hoc queries because they do not reuse cached plan as good as stored procedures do especially if Ad-Hoc queries work with parameters, so in this context, we might find many cached plans for only one Ad-Hoc query and it will then make more difficult spot what we are looking for. It is of paramount importance to remember that optimisation and tuning is a never-ending task and it is worthy of our attention every so often. Today I am coming with the following script to help quickly find a cached plan execution plan for an ad-hoc query. This query filters out by the specific text you are looking for in the ad-query.

SELECT TOP 10 st.text AS [SQLText], cp.cacheobjtype CacheObjType, cp.objtype ObjType,
COALESCE(DB_NAME(st.dbid), DB_NAME(CAST(pa.value AS INT))+'*',
'Resource') AS [DatabaseName], cp.usecounts AS [PlanUsage], qp.query_plan
FROM sys.dm_exec_cached_plans cp
CROSS APPLY sys.dm_exec_sql_text(cp.plan_handle) st
CROSS APPLY sys.dm_exec_query_plan(cp.plan_handle) qp
OUTER APPLY sys.dm_exec_plan_attributes(cp.plan_handle) pa
WHERE pa.attribute = 'dbid'
AND st.text LIKE '%some text of the object or ad-hoc query%'

It well worth noting that this is why it is highly advisable using stored procedures so that CPU and memory resources can be used more efficiently. That is all for now, let me know any remarks you may have. Stay tuned.

Tuesday 12 September 2017

Detecting poor cursor usage in SQL Server

Without a shadow of a doubt, poor cursor usage issue is an essential aspect to review so as to reduce the possibility of having CPU bottleneck issues, and determine whether cursors are the most appropriate means to accomplish the processing or whether a set-based operation is possible. It is well know that set-based operation is generally more efficient, but if you make the decision of using cursors, you should ensure that it does not represent an issue for the database in question. Thus, detecting poor cursor usage and taking certain measures to ease the problem is crucial, and in SQL Server there are some techniques available to be used to diagnose the issue. To begin with, by using performance counters we are able to analyse the issue and find out the extent to which poor cursor usage is adversely affecting on the performance of specific workloads or the whole database. For instance, here is the counter SQL Server: Cursor Manager By Type: Cursor Requests/Sec which retrieves information about the number of SQL cursor requests received by the server. On the top of that, it is possible to filter out by the cursor manager instance such as API Cursor (only the API cursor information), TSQL Global Cursor and TSQL Local Cursor. Have a look at SQL Server, Cursor Manager by Type Object to get more info about the counter.
On the other hand, using SQL Trace is also helpful, for example, use a trace that includes the RPC:Completed event class search for 'sp_cursorfetch' statements. The value of the fourth parameter is the number of rows returned by the fetch. It is worth noting that the maximum number of rows that are requested to be returned is specified as an input parameter in the corresponding RPC:Starting event class. Finally, by using the DMV 'sys.dm_exec_cursors' we can also determine whether poor cursor usage exists in the database server as Transact-SQL cursors always have a fetch buffer of 1 and for API cursors it should be higher.

select c.* 
from sys.dm_exec_sessions s
    cross apply sys.dm_exec_cursors(con.session_id) as c
where cur.fetch_buffer_size = 1 
    and cur.properties LIKE 'API%'

Consequently, if it is seen that API Cursors have a fetch buffer size of 1 then consider enabling multiple active results (MARS) when connecting to SQL Server and consult the appropriate documentation for your specific API to determine how to specify a higher fetch buffer size for the cursor either ODBC (SQL_ATTR_ROW_ARRAY_SIZE) or OLE DB (IRowset::GetNextRows, IRowsetLocate::GetRowsAt). After that, we can retrieve more details about the session and connection of the users associated to the harmful cursors so as to decide what to do.

select s.session_id, cn.client_net_address, s.login_name, s.status,s.client_interface_name, s.program_name, 
       c.cursor_id, c.name, c.properties, c.plan_generation_num, c.creation_time, c.is_open, c.fetch_status, 
       c.fetch_buffer_size, c.worker_time, c.reads, c.writes, c.dormant_duration
from sys.dm_exec_connections cn
inner join  sys.dm_exec_sessions s on cn.session_id = s.session_id
cross apply sys.dm_exec_cursors(s.session_id) as c
where c.fetch_buffer_size = 1 
     and c.properties LIKE 'API%'

That is all for now, thanks for reading. Let me know any remarks you may have. Stay tuned.

Thursday 7 September 2017

Getting information about memory used by SQL Server

Naturally, every DBA is asked to report information about the memory used by SQL Server. Common questions are related to memory reserved and memory used currently, and sometimes we might not know how to complete this task. How many times did we take a look at the Windows Manager Task to find out that info? Many of us might have ended up quite frustrated time and time again because it did not help much. Over time Microsoft decided to provide Administrators with more useful tools and released Resource Manager tool whereby useful memory information per process is available, nevertheless, whether we want to get that info from SQL Server it might be an uphill battle as it would need that we write certain complex code at windows level.  For the time being, thinking about this situation I made the snap decision of sharing with you some helpful scripts that will alleviate the pain.

In the likely event that you might need to get the total buffer pool memory used by all databases at SQL instance level, this script is for it.

SELECT cast( cast( COUNT(*) /128.0/1024.0 as decimal(10,2)) as varchar(10)) + 'GB'  AS TotalUsageBufferPool
FROM sys.dm_os_buffer_descriptors

And if you wanted to know the buffer pool memory used by each database, you can use this:

SELECT CASE database_id 
        WHEN 32767 THEN 'ResourceDb'  ELSE db_name(database_id)         END AS DatabaseName,
        cast( COUNT(*) /128.0 as decimal(10,2)) AS [BufferPool(MB)]
FROM sys.dm_os_buffer_descriptors
GROUP BY DB_NAME(database_id) ,database_id
ORDER BY [BufferPool(MB)] DESC

Finally, and more importantly, answering the question about the total memory used by the whole SQL instance:
-- SQL2012/2014/2016/2017
select cast(cast(physical_memory_kb /1024.0/1024.0  as decimal(10,2))  as varchar(10)) + 'GB' TotalPhysicalRAM, 
cast(cast(visible_target_kb /1024.0/1024.0  as decimal(10,2)) as varchar(10)) + 'GB' MaxRAM, -- max memory configure at sql server level
cast(cast(committed_target_kb /1024.0/1024.0  as decimal(10,2)) as  varchar(10)) + 'GB' ReservedRAM,  --memory reserved
cast(cast(committed_kb /1024.0/1024.0  as decimal(10,2)) as  varchar(10)) + 'GB' UsedRAM --memory used currently
FROM sys.dm_os_sys_info

-- for SQL2005/2008/2008R2
SELECT cast(cast(physical_memory_in_bytes /1024.0/1024.0  as decimal(10,2))  as varchar(10)) + 'GB' TotalPhysicalRAM, 
cast(cast(bpool_visible /128.0/1024.0  as decimal(10,2)) as varchar(10)) + 'GB' MaxRAM, -- max memory configure at sql server level
cast(cast(bpool_commit_target /128.0/1024.0  as decimal(10,2)) as  varchar(10)) + 'GB' ReservedRAM,  --memory reserved
cast(cast(bpool_committed /128.0/1024.0  as decimal(10,2)) as  varchar(10)) + 'GB' UsedRAM --memory used currently
FROM sys.dm_os_sys_info

That is all for now. I hope you find these scripts helpful. Let me know any remarks you may have.
HELLO, I'M PERCY REYES! — a book lover, healthy lifestyle lover... I've been working as a senior SQL Server Database Administrator (DBA) for over 20 years; I'm a three-time awarded Microsoft Data Platform MVP. I'm currently doing a PhD in Computer Science (cryptography) at Loughborough University, England — working on cryptographic Boolean functions, algorithmic cryptanalysis, number theory, and other algebraic aspects of cryptography. READ MORE