Scribd
Upload
298 views

SQL Server Performance Tuning

The document provides an overview of options for tuning the performance of a SQL Server database without adding additional hardware, including adjusting queries and indexes, optimizing the procedure cache, understanding table and index sizes, and tuning server settings. It also lists resources available on the author's website for downloading queries, procedures, and a full presentation on SQL Server performance tuning techniques.

Uploaded by

Bang Trinh
Copyright
© © All Rights Reserved
Available Formats
Download as PDF, TXT or read online on Scribd
298 views

SQL Server Performance Tuning

The document provides an overview of options for tuning the performance of a SQL Server database without adding additional hardware, including adjusting queries and indexes, optimizing the procedure cache, understanding table and index sizes, and tuning server settings. It also lists resources available on the author's website for downloading queries, procedures, and a full presentation on SQL Server performance tuning techniques.

Uploaded by

Bang Trinh
Copyright
© © All Rights Reserved
Available Formats
Download as PDF, TXT or read online on Scribd
You are on page 1/ 25

Steve Stedman – Programmer and Database Consultant

Email: [email protected]
 General Scaling Options
 General Performance Tuning
 Indexes
 Query Tuning
 Procedure Cache Tuning
 Understanding Table Sizes

 All referenced queries and stored procedures are


available at http://SteveStedman.com/sql
 Purchase a larger server, and replace the
existing system.
 Works well with smaller systems.
 Cost prohibitive for larger systems.
 Can be a temporary solution.
 Purchase more hardware and split or partition
the database.
 Partitioning can be either vertical or
horizontal.
◦ Vertical: Split the databases based on a specific
demographic such as time zone or zip code.
◦ Horizontal: Split components out of one database
into another.
 Adjusting and rewriting queries.
 Adding indexes.
 Removing indexes.
 Re-architecting the database schema.
 Moving things that shouldn‟t be in the
database.
 Eliminating redundant work on the database.
 Caching of data.
 Other performance tuning techniques.
 Bigger hardware?
 More hardware?
 Tuning without adding more hardware?

 There is a time and a place for each of these


depending on your specific environment.
 This presentation covers tuning without
adding more hardware.
 Allowing your system to scale.
◦ Adding more customers.
◦ Adding more features.
 Improve overall system performance.
 Save money by not wasting resources.
◦ The database is typically one of the most expensive
resources in a datacenter. Make the most of it.
 Non-Clustered Indexes
◦ Traditional Indexing – contains pointers to the data.

 Clustered Indexes
◦ Reorganizes the actual data on disk.

 Covered Indexes
◦ Returns query results without accessing the base
table.
◦ Can lead to major performance increases.
◦ Applies to Non-Clustered Indexes.
 Contain only the data specified in the index.
 Do not change the base layout of the tables.
 Use pointers to get to the data.
 Can be created on most data types including
char(), varchar(), and uniqueidentifiers.
 Only one non-clustered index can be used
per table reference in a query.
 Can improve performance with multiple
columns.
 Causes base table structure to change.
 Only one clustered index per table.
 Should never contain char(), varchar(),
varbinary(), uniqueidentifiers, or other large
or widely distributed identifiers.
 Can significantly increase the size of a table
and the database.
 Can increase performance.
Table Name Rows Disk Space Index Space Total Space
(MB) (MB) (MB)
Table1 2,034,998 450 872 1,322

Table2 2,423,745 476 639 1,115

Table3 2,194,080 318 656 974

Table4 396,241 598 60 658

Table5 1,812,884 223 371 595

Use the DiskUsageByTable stored procedure to access these numbers.


 Find out which indexes are being used and
which are not.
SELECT o.name AS object_name, i.name AS index_name,
i.type_desc, u.user_seeks, u.user_scans,
u.user_lookups, u.user_updates
FROM sys.indexes i
JOIN sys.objects o ON i.object_id = o.object_id
LEFT JOIN sys.dm_db_index_usage_stats u ON i.object_id = u.object_id
AND i.index_id = u.index_id
AND u.database_id = DB_ID()
WHERE o.type <> 'S' -- No system tables!
ORDER BY (ISNULL(u.user_seeks, 0) + ISNULL(u.user_scans, 0) + ISNULL(u.user_lookups, 0) +
ISNULL(u.user_updates, 0)), o.name, i.name

 If indexes are not being used, find out why, and if


they are really needed. If they are not needed, then
remove them.
 An Index Scan accesses all the rows in the
index.
 An Index Seek uses selective rows in the
index.

 The Seek is much quicker than the scan.


 Avoid SELECT * FROM … instead select
specific column names. Just ask for the
columns you are looking for.

 Using LIKE clause


◦ WHERE ____ LIKE „ste%‟ uses indexes.
◦ WHERE ____ LIKE „%son‟ cannot use an index.
 WHERE age + 5 > 65
◦ Instead use WHERE age > 65 – 5

 WHERE ISNULL(order_date, „Jan 01,2003‟) >


„Jan 01, 2002 12:00:00 AM‟
◦ Instead use WHERE ord_date IS NOT NULL
AND ord_date > „Jan 01, 2002 12:00:00 AM‟
 When running queries through the SQL Server
Management Studio, turn on “Include Actual
Execution Plan” from the Query Menu.
 SQL Server 2008 will give recommendations
on missing indexes.
 The procedure cache caches more than just
procedures, it also caches parsed queries.
 Performance tuning the Procedure Cache reduces
waste on the SQL Server.
 You don‟t have control over the size of the
procedure cache, but you do have control over
how it is used.
 Reuse in the procedure cache allows queries and
procedures to run faster.
 Tuning is accomplished by a number of methods:
◦ Using Parameterized Queries.
◦ Removing temp tables from Procedures.
◦ Implementing database Coding Standards
 You can examine the queries in the procedure
cache with the following query.

SELECT qs.execution_count,
st.text, total_elapsed_time
FROM sys.dm_exec_query_stats AS qs
CROSS APPLY sys.dm_exec_sql_text(qs.sql_handle) AS st
WHERE st.encrypted = 0
ORDER BY st.text

 Scroll through results and find similar queries


taking up space in the procedure cache.
 Allows for already parsed queries to be re-used.
 Replace hard coded values with parameters.
SQLCommand cmd = new SqlCommand("SELECT column1 from table_name
where column2 = „Customer Name‟", conn);
reader = cmd.ExecuteReader();

 Replace with

SQLCommand cmd = new SqlCommand("SELECT column1 from table_name


where column2 = @PARAM1", conn);
SqlParameter param = new SqlParameter();
param.ParameterName = "@PARAM1";
param.Value = “Customer Name”;
cmd.Parameters.Add(param);
reader = cmd.ExecuteReader();
 A One Time Use Query is a query that is
compiled in procedure cache run once and
never used again.
 If a query takes up 5mb of memory just for
the parsing, this can add up quickly if there
are dozens or hundreds of instances in
memory.
 These can be reduced with parameterization.
 Use the stored proc FindOneTimeUseQueries
 The table size is determined by the following:
◦ Data in the table.
◦ Clustered indexes on the table.
◦ Non-Clustered indexes on the table.
 DiskUsageByTable stored procedure
 Don‟t let one table take
over your whole database.
 Focus on the problems.
 Turn off any unneeded services.
◦ FTP
◦ Webserver
◦ Others…
 “Maximize Throughput for network
Applications”.
 Turn off any screensavers.
 There are many ways to performance tune
your SQL Server. You can use any or all of the
methods outlined in this presentation.
 Determine which method is best for your
needs:
◦ Tuning Queries
◦ Tuning Indexes
◦ Tuning the Procedure Cache
◦ Analyzing Table and Index Sizes
◦ Adding Hardware
 Visit my website:
◦ Download this presentation, stored procedures and
queries:
◦ http://SteveStedman.com

 Contact Me:
◦ [email protected]

You might also like