Monthly Archives: September 2013

SQL Server Storage Engine: Database Files and Filegroups

Recently I have received a few emails asking me to clarify a few things from the old blog posts I wrote way back in 2010. After I re-read those posts, I decided that it could make sense to refresh and rewrite some of them. I hope, it can be done better this time. 🙂

In the next a few months I will talk a bit about SQL Server Storage Engine covering how SQL Server stores the data; what is the format of data row and data page; what are the allocation maps; and so on. We will see how it goes and where to stop.

Today I will start writing a few words about SQL Server database files and filegroups in general.

SQL Server database is a collection of the objects that allow us to store and manipulate the data. In theory, SQL Server supports 32,767 databases per instance although the typical installation usually has just several databases. Obviously, the number of the databases SQL Server can handle depends on the load and hardware. It is not unusual to see the servers hosting dozens or even hundreds of small databases.

Every database consists of one or more transaction log and one or more data files. Transaction log stores the information about database transactions and all data modifications made by each session. Every time the data has been modified, SQL Server stores enough information in the transaction log to undo (rollback) or redo (replay) the action.

Every database has one primary data file, which, by default, has .mdf extension. In addition, every database can have secondary database files. Those files, by default, have .ndf extension.

All database files are grouped into the filegroups. Filegroup is the logical unit, which simplifies database administration. They allow the separation between logical object placement and physical database files. When you create the database objects-tables, for example-you specify in what filegroup they should be placed without worrying about underlying data files configuration.

The script shown below creates the database with name OrderEntryDb. That database consists of three filegroups. The primary filegroup has one data file stored on M: drive. Second filegroup- Entities– has one data file on N: drive. Last filegroup- Orders– has two data files stored on O: and P: drives. Finally, there is the transaction log file on L: drive.

create database [OrderEntryDb] on 
(name = N'OrderEntryDb', filename = N'm:\OEDb.mdf'),
filegroup [Entities] 
(name = N'OrderEntry_Entities_F1', filename = N'n:\OEEntities_F1.ndf'),
filegroup [Orders] 
(name = N'OrderEntry_Orders_F1', filename = N'o:\OEOrders_F1.ndf'),
(name = N'OrderEntry_Orders_F2', filename = N'p:\OEOrders_F2.ndf') 
log on
(name = N'OrderEntryDb_log', filename = N'l:\OrderEntryDb_log.ldf')

You can see the physical layout of the database and data files below. There are five disks with four data- and one transaction- log files. Dashed rectangles represent the filegroups.

01. Files and Filegroups

Ability to put multiple data files inside the filegroup allows us to spread the load across different storage devices, which would help to improve I/O performance of the system. Transaction log, on the other hand, does not benefit from the multiple files. SQL Server works with transaction log in sequential matter and multple log files just stay idle.

Let’s create a few tables in the database we created. The tables Clients and Articles are placed into Entities filegroup. The table Orders resides in Orders filegroup.

create table dbo.Customers
    -- Table columns
) on [Entities];

create table dbo.Articles
    -- Table columns
) on [Entities];

create table dbo.Orders
    -- Table columns
) on [Orders];

The physical layout of the tables in the database and disks is shown below.

02. Tables and Filegroups

The separation between logical object placement in the filegroups and physical database files allow us to fine-tune the database file layout getting the most from the storage subsystem. For example, independent software vendors (ISV), who are deploying their products to different customers, can adjust the number of database files based on underlying I/O configuration and expected amount of the data during deployment stage. Those changes would be transparent to the developers, who are placing the database objects to the filegroups rather than database files.

It is generally recommended to avoid using PRIMARY filegroup for anything but system objects. Creating separate filegroup or set of the filegroups for the user objects simplifies database administration and disaster recovery especially in case of the large databases.

You can specify initial file size and auto-growth parameters at time when you create the database or add new files to existing database. SQL Server uses proportional fill algorithm when choosing in what data file it should write data to. It writes an amount of data proportionally to the free space available in the files – more free space are in the file, more writes it would handle.

I would recommend that all files in the single filegroup would have the same initial size and auto-growth parameters with grow size defined in megabytes rather than percent. This would help proportional fill algorithm evenly balance write activities across data files.

Every time SQL Server grows the files, it fills newly allocated space in the files with zeros. That process blocks all sessions that need to write to the corresponding file or, in case of transaction log growth, generate transaction log records.

SQL Server always zeroing out transaction log and that behavior cannot be changed. Although, you can control if data files are zeroing out or not by enabling or disabling Instant File Initialization. Enabling Instant File Initialization helps to speed up data file growth and reduces the time required to create or restore the database.

There is the small security risk associated with Instant File Initialization. When this option is enabled, unallocated part of the data file can contain the information from the previously deleted OS files. Database administrators will be able to examine such data.

You can enable Instant File Initialization by adding SA_MANAGE_VOLUME_NAME permission also known as “Perform Volume Maintenance Task” to SQL Server startup account. This can be done under Local Security Policy management application (secpol.msc) as shown below. You need to open properties for “Perform volume maintenance task” permission and add SQL Server startup account to the list of users there.

03. Instant File Initialization: Local Security Policy

SQL Server checks if it has Instant File Initialization enabled on startup. You need to restart SQL Server service after you add corresponding permission. 

In order to check if permission is enabled, you can use the code from the listing below. This code sets two trace flags that forces SQL Server to put  additional information to the error log, creates the small database and reads the content of the log.

-- add more output to error log
dbcc traceon(3004,3605,-1)
create database Dummy
exec sp_readerrorlog
drop database Dummy
dbcc traceoff(3004,3605,-1)

In case, if Instant File Initialization is not enabled, SQL Server error log shows that SQL Server zeroing mdf data file in addition to zeroing log .ldf file as shown below. When Instant File Initialization is enabled, it would only mention zeroing of the log .ldf file.

04. Instant File Initialization: Checking if instant file initialization is enabled

Another important database option that controls the database file sizes is Auto Shrink. When this option is enabled, SQL Server regularly shrinks the database files, reduces their size and release space to operating system. This operation is very resource intensive and rarely useful – the database files grow up again after some time when new data comes to the system. Auto Shrink must never be enabled on the database. Moreover, Microsoft would remove that option in the future versions of the SQL Server.

Next: Data Pages and Data Rows

Table of Content

Slide decks and a few announcements

I am speaking at SQL Saturday #190 in Denver, CO on Saturday, September 28th, 2013. I’m also delivering full day pre-con on Friday, September 27th. We will talk about SQL Server Internals and cover a lot of things – internal storage, locking, troubleshooting, internal implementation of the database objects and so on. There are still a few seats available and I will be happy to see you there.

Speaking of the past presentations – the slide decks and demo scripts for both parts of “All About Indexes” presentations are available for download from the Presentations page. Keep in mind, that slides are in Russian.

The slide deck for “From S to 3XL: Implementing Data Partitioning and Tiered Storage” session will be available for download after PASS SUMMIT.


Optimizing SQL Server Spatial Queries with Bounding Box

SQL Server supports two data types to store spatial information – geometry and geography. Geometry supports planar, or Euclidean, flat-earth data. Geography supports ellipsoidal round-earth surface. Both data types can be used to store location information, such as GPS latitude and longitude coordinates. Geography data type considers Earth roundness and provides slightly better accuracy although it has stricter requirements to the data. As a couple of examples, data must fit in the single hemisphere and polygons must be defined in specific ring orientation.

Storing location information in geometry data type introduces its own class of problems. It works fine and often has better performance than geography data type. Although, we cannot calculate the distance between points – the unit of measure for result is decimal degrees, which are useless in non-flat surface.

Let’s take a look at spatial data type performance in one of the specific use-cases, such as distance calculation between two points. Typical use-case for that scenario is the search for the point of interest (POI) close to specific location. First, let’s create three different tables storing POI information in the different format and populate them with some data.

01. Test tables

02. Populating test tables with data

The first table dbo.Locations stores the coordinates using decimal(9,6) data type. Two other tables are using geography data type. Finally, the table dbo.LocationsGeoIndexed has Location column indexed with special type of the index called spatial index. Those indexes help improving performance of some operations, for example distance calculation or check if objects are intersecting.

It is worth mentioning that the first table uses decimal(9,6) data type rather than float. This decision saves us six bytes per pair of values and provides the accuracy that exceed the accuracy of commercial-grade GPS receivers.

Let’s run the tests that measures performance of the queries that calculate the number of locations within one mile from Seattle City Center. In case of dbo.Locations table, we will use dbo.CalcDistanceCLR function, which we defined earlier. For two other tables we will call spatial method STDistance.

03. Test queries (table-wide lookup)

04. Execution plans (table-wide lookup)

The first and second queries perform clustered index scan and calculate the distance for every row from the table. Last query uses spatial index to lookup such rows. We can see execution times for the queries in the table below.

05. Execution time (Table-wide lookup)

As we see, spatial index greatly benefits the query. It is also worth mentioning, that without the index, performance of CalcDistanceCLR method is better comparing to STDistance.

Although spatial index greatly improves the performance, it has its own limitations. It works in the scope of entire table and all other predicates are evaluated after spatial index operations. That can introduce suboptimal plans in some cases.

As the example, let’s look at the use-case, when we store POI information on company-by-company basis .

06. Test table creation (company-wide lookup)

In case, when we perform POI lookup for specific company, CompanyId column must be included as the predicate to the queries. SQL Server has two choices how to proceed. The first choice is clustered index seek based on CompanyId value calling STDistance method for every POI that belongs to the company. Another choice is using spatial index, find all POIs within the specified distance regardless of what company they belong and, finally, join it with the clustered index data. Let’s run those queries.

07. Test queries (company-wide lookup)

Neither method is efficient in case when table stores the large amount of data for the large number of companies. Execution plan of the first query utilizing clustered index seek shows that it performed STDistance call 241,402 times – once per every company POI.

08. Execution plan (clustered index seek approach)

The execution plan for the second query shows that spatial index lookup returned 550 rows – all POI in the area, regardless of what company they belong. SQL Server had to join the rows with the clustered index before evaluating CompanyId predicate.

09. Execution plan (Spatial index approach)

One of the ways to solve such problem called Bounding Box approach. That method allows us to minimize the number of the calculations by filtering out POIs that are outside of the area of interest.

10. Bounding box

All points we need to select residing in the circle with location as the center point and radius specified by the distance. The only points we need to evaluate are residing within the box that inscribes that circle.

We can calculate the coordinates of the corner points of the box, persist it in the table and use regular non-clustered index to pre-filter the data. This would allow us to minimize the number of expensive distance calculations to perform.

Calculation of the bounding box corner points can be done with CLR table-valued function shown below.

11. Calculating bounding box corner points

Let’s alter our table and add bounding box points. We also need to create non-clustered index to support our query.

12. Table alteration (adding bounding box corner points)

Now, we can change the query to utilize the bounding box.

13. Query utilizing bounding box

14. Query utilizing bounding box (Execution plan)

As we see, last query calculated the distance 15 times. This is significant improvement comparing to 241,402 calculations from the original query. The execution times are shown below:

15. Execution time (Company-wide lookup)

As we see, bounding box outperforms both – clustered index seek and spatial index lookup. Obviously, it would be the case only when bounding box reduces the number of the calculations to degree that offsets the overhead of non-clustered index seek and key lookup operations. It is also worth mentioning that we do not need spatial index with such approach.

We can use bounding box for the other use-cases. For example, when we are checking if position belongs to the area defined by the polygon. Bounding box corner coordinates should store minimum and maximum latitude/longitude of the polygon corner points. Similarly to the distance calculation, we would filter-out the locations outside of the box before performing expensive spatial method call that validates if point is within the polygon area.

Source code is available for download