Редагувати

Поділитися через


Create partitioned tables and indexes

Applies to: SQL Server Azure SQL Database Azure SQL Managed Instance

You can create a partitioned table or index in SQL Server, Azure SQL Database, and Azure SQL Managed Instance by using SQL Server Management Studio or Transact-SQL. The data in partitioned tables and indexes is horizontally divided into units that can be spread across more than one filegroup in a database, or stored in a single filegroup. Partitioning can make large tables and indexes more manageable and scalable.

Creating a partitioned table or index typically happens in three or four parts:

  1. Optionally create a filegroup or filegroups and corresponding data files that will hold the partitions specified by the partition scheme. The main reason to place partitions on multiple filegroups is to ensure you can independently perform backup and restore operations on filegroups. If this isn't required, you might choose to assign all partitions to a single filegroup, using either an existing filegroup, such as PRIMARY, or a new filegroup with related data files. In nearly all scenarios, you'll achieve all benefits of partitioning whether or not you use multiple filegroups.

  2. Create a partition function that maps the rows of a table or index into partitions based on the values of a specified column. You can use a single partition function to partition multiple objects.

  3. Create a partition scheme that maps the partitions of a partitioned table or index to one filegroup or to multiple filegroups. You can use a single partition scheme to partition multiple objects.

  4. Create or alter a table or index and specify the partition scheme as the storage location, along with the column that will serve as the partitioning column.

Note

Partitioning is fully supported in Azure SQL Database. Because only the PRIMARY filegroup is supported in Azure SQL Database, all partitions must be placed on the PRIMARY filegroup.

Table partitioning is also available in dedicated SQL pools in Azure Synapse Analytics, with some syntax differences. Learn more in Partitioning tables in dedicated SQL pool.

Permissions

Creating a partitioned table requires CREATE TABLE permission in the database and ALTER permission on the schema in which the table is being created. Creating a partitioned index requires ALTER permission on the table or view where the index is being created. Creating either a partitioned table or index requires any one of the following additional permissions:

  • ALTER ANY DATASPACE permission. This permission defaults to members of the sysadmin fixed server role and the db_owner and db_ddladmin fixed database roles.

  • CONTROL or ALTER permission on the database in which the partition function and partition scheme are being created.

  • CONTROL SERVER or ALTER ANY DATABASE permission on the server of the database in which the partition function and partition scheme are being created.

Create a partitioned table on one filegroup using Transact-SQL

If you don't need to independently perform backup and restore operations on filegroups, partitioning a table using a single filegroup simplifies management of the partitioned table over time.

This example is suitable for Azure SQL Database, which doesn't support adding files and filegroups. Table partitioning is supported in Azure SQL Database by creating partitions in the PRIMARY filegroup. For SQL Server and Azure SQL Managed Instance, you might wish to specify a user-created filegroup, depending on your filegroup and file management practices.

The example steps through creating a partitioned table in SQL Server Management Studio (SSMS) using Transact-SQL and assigns all partitions to the PRIMARY filegroup. The example:

  • Creates a RANGE RIGHT partition function named myRangePF1 with three boundary values using the datetime2 data type. Three boundary values will result in a partitioned table with four partitions.
  • Creates a partition scheme named myRangePS1 that uses the ALL TO syntax to assign all partitions in the myRangePF1 partition function to the PRIMARY filegroup.
  • Creates a table named PartitionTable on the myRangePS1 partition scheme, specifying a column named col1 as the partitioning column.
  1. In Object Explorer, connect to an instance of Database Engine.

  2. On the Standard bar, select New Query.

  3. Copy and paste the following example into the query window and select Execute. This example creates a partition function and a partition scheme. A new table is created with the partition scheme specified as the storage location.

CREATE PARTITION FUNCTION myRangePF1 (datetime2(0))
    AS RANGE RIGHT FOR VALUES ('2022-04-01', '2022-05-01', '2022-06-01') ;
GO

CREATE PARTITION SCHEME myRangePS1
    AS PARTITION myRangePF1
    ALL TO ('PRIMARY') ;
GO

CREATE TABLE dbo.PartitionTable (col1 datetime2(0) PRIMARY KEY, col2 char(10))
    ON myRangePS1 (col1) ;
GO

Create a partitioned table on multiple filegroups with Transact-SQL

Follow the steps in this section to create one or more filegroups, corresponding files, and a partitioned table using Transact-SQL in SSMS.

Both SQL Server and Azure SQL Managed Instance support creating filegroups and files. Azure SQL Managed Instance automatically configures the path for all database files added, so the ALTER DATABASE ADD FILE command in Azure SQL Managed Instance doesn't allow the FILENAME parameter. Azure SQL Database supports creating partitioned tables only in the PRIMARY filegroup. Find example code for Azure SQL Database in Create a partitioned table on one filegroup using Transact-SQL.

Run the following example against an empty database. The example:

  • Adds four new filegroups to a database.
  • Adds one file to each filegroup.
  • Creates a RANGE RIGHT partition function called myRangePF1 with three boundary values that will partition a table into four partitions.
  • Creates a partition scheme called myRangePS1 that applies myRangePF1 to the four new filegroups.
  • Creates a partitioned table called PartitionTable that uses myRangePS1 to partition col1.
  1. In Object Explorer, connect to an instance of Database Engine.

  2. On the Standard toolbar, select New Query.

  3. This example creates a new database and uses it. It then creates new filegroups, a partition function, and a partition scheme. A new table is created with the partition scheme specified as the storage location. Copy and paste the following example into the query window.

    If you're using a managed instance, remove the FILENAME parameter and associated value from the ALTER DATABASE ADD FILE command. The managed instance determines the file path for you automatically.

    If you're using a SQL Server instance, customize the value for the FILENAME parameter to a location appropriate for your instance.

    If you wish to use an existing database, remove the CREATE DATABASE command and alter the USE statement to the appropriate database name.

    SELECT Execute.

    CREATE DATABASE PartitionTest;
    GO
    
    USE PartitionTest;
    GO
    
    ALTER DATABASE PartitionTest
    ADD FILEGROUP test1fg;
    GO
    ALTER DATABASE PartitionTest
    ADD FILEGROUP test2fg;
    GO
    ALTER DATABASE PartitionTest
    ADD FILEGROUP test3fg;
    GO
    ALTER DATABASE PartitionTest
    ADD FILEGROUP test4fg;
    
    ALTER DATABASE PartitionTest
    ADD FILE
    (
        NAME = partitiontest1,
        FILENAME = 'C:\Program Files\Microsoft SQL Server\MSSQL15.MSSQLSERVER\MSSQL\DATA\partitiontest1.ndf',
        SIZE = 5MB,
        FILEGROWTH = 5MB
    )
    TO FILEGROUP test1fg;
    ALTER DATABASE PartitionTest
    ADD FILE
    (
        NAME = partitiontest2,
        FILENAME = 'C:\Program Files\Microsoft SQL Server\MSSQL15.MSSQLSERVER\MSSQL\DATA\partitiontest2.ndf',
        SIZE = 5MB,
        FILEGROWTH = 5MB
    )
    TO FILEGROUP test2fg;
    GO
    ALTER DATABASE PartitionTest
    ADD FILE
    (
        NAME = partitiontest3,
        FILENAME = 'C:\Program Files\Microsoft SQL Server\MSSQL15.MSSQLSERVER\MSSQL\DATA\partitiontest3.ndf',
        SIZE = 5MB,
        FILEGROWTH = 5MB
    )
    TO FILEGROUP test3fg;
    GO
    ALTER DATABASE PartitionTest
    ADD FILE
    (
        NAME = partitiontest4,
        FILENAME = 'C:\Program Files\Microsoft SQL Server\MSSQL15.MSSQLSERVER\MSSQL\DATA\partitiontest4.ndf',
        SIZE = 5MB,
        FILEGROWTH = 5MB
    )
    TO FILEGROUP test4fg;
    GO
    
    CREATE PARTITION FUNCTION myRangePF1 (datetime2(0))
        AS RANGE RIGHT FOR VALUES ('2022-04-01', '2022-05-01', '2022-06-01') ;
    GO
    
    CREATE PARTITION SCHEME myRangePS1
        AS PARTITION myRangePF1
        TO (test1fg, test2fg, test3fg, test4fg) ;
    GO
    
    CREATE TABLE PartitionTable (col1 datetime2(0) PRIMARY KEY, col2 char(10))
        ON myRangePS1 (col1) ;
    GO
    

Partition a table with SSMS

Follow the steps in this section to optionally create filegroups and corresponding files, then create a partitioned table or partition an existing table using the Create Partition Wizard in SQL Server Management Studio (SSMS). The Create Partition Wizard is available in SSMS for SQL Server and Azure SQL Managed Instance. For Azure SQL Database, refer to Create a partitioned table on one filegroup using Transact-SQL.

Create new filegroups (optional)

If you wish to place your partitioned table on one or more new filegroups, follow the steps in this section. Both SQL Server and Azure SQL Managed Instance support creating filegroups and files. For Azure SQL Managed Instance, the path for any files created will automatically be configured for you.

  1. In Object Explorer, right-click the database in which you want to create a partitioned table and select Properties.

  2. In the Database Properties - database_name dialog box, under Select a page, select Filegroups.

  3. Under Rows, select Add. In the new row, enter the filegroup name.

    Warning

    When specifying multiple filegroups, you must always have one extra filegroup in addition to the number of filegroups specified for the boundary values when you are creating partitions.

  4. Continue adding rows until you have created all of the filegroups for the partitioned table or tables.

  5. Select OK.

  6. Under Select a page, select Files.

  7. Under Rows, select Add. In the new row, enter a filename and select a filegroup.

  8. Continue adding rows until you have created at least one file for each filegroup.

Create a partitioned table

  1. Optionally, expand the Tables folder and create a table as you normally would. For more information, see Create tables (Database Engine). Alternatively, you can specify an existing table in the next step.

  2. Right-click the table that you wish to partition, point to Storage, and then select Create Partition....

  3. In the Create Partition Wizard, on the Welcome to the Create Partition Wizard page, select Next.

  4. On the Select a Partitioning Column page, in the Available partitioning columns grid, select the column on which you want to partition your table. Only columns with data types that can be used to partition data will be displayed in the Available partitioning columns grid. If you select a computed column as the partitioning column, the column must be created as a persisted column.

    The choices you have for the partitioning column and the values range are determined primarily by the extent to which your data can be grouped in a logical way. For example, you might choose to divide your data into logical groupings by months or quarters of a year. The queries you plan to make against your data determine whether this logical grouping is adequate for managing your table partitions. All data types are valid for use as partitioning columns, except text, ntext, image, xml, timestamp, varchar(max), nvarchar(max), varbinary(max), alias data types, or common language runtime (CLR) user-defined data types.

    The following additional options are available on this page:

    Collocate this table to the selected partitioned table
    Allows you to select a partitioned table that contains related data to join with this table on the partitioning column. Tables with partitions joined on the partitioning columns are typically queried more efficiently.

    Storage-align non-unique indexes and unique indexes with an indexed partition column
    Aligns all indexes of the table that are partitioned with the same partition scheme. When a table and its indexes are aligned, you can move partitions in and out of partitioned tables more effectively, because your data is partitioned in the same way.

    After selecting the partitioning column and any other options, select Next.

  5. On the Select a Partition Function page, under Select partition function, select either New partition function or Existing partition function. If you choose New partition function, enter the name of the function. If you choose Existing partition function, select the name of the function you'd like to use from the list. The Existing partition function option won't be available if there are no other partition functions on the database.

    After completing this page, select Next.

  6. On the Select a Partition Scheme page, under Select partition scheme, select either New partition scheme or Existing partition scheme. If you choose New partition scheme, enter the name of the scheme. If you choose Existing partition scheme, select the name of the scheme you'd like to use from the list. The Existing partition scheme option won't be available if there are no other partition schemes on the database.

    After completing this page, select Next.

  7. On the Map Partitions page, under Range, select either Left boundary or Right boundary. Left boundary specifies that the highest bounding value will be included within a partition. Right boundary specifies that the lowest bounding value will be included in each partition. Learn more about right and left ranges in Partition function.

    When specifying multiple boundary points, you must always enter one extra row in addition to the rows assigning boundary values to a filegroup.

    In the Select filegroups and specify boundary values grid, under Filegroup, select the filegroup into which you want to partition your data. Under Boundary, enter the boundary value for each filegroup. If you wish to assign multiple or all partitions to the same filegroup, select the same filegroup name for each row. If you select a filegroup on a single row and boundary value is left empty, the partition function maps the whole table or index into a single partition using the partition function name.

    The following additional options are available on this page:

    Set Boundaries...
    Opens the Set Boundary Values dialog box to select the boundary values and date ranges you want for your partitions. This option is only available when you have selected a partitioning column that contains one of the following data types: date, datetime, smalldatetime, datetime2, or datetimeoffset.

    Estimate storage
    Estimates rowcount, required space, and available space for storage for each filegroup specified for the partitions. These values are displayed in the grid as read-only values.

    The Set Boundary Values dialog box allows for the following additional options:

    Start date
    Selects the starting date for the range values of your partitions.

    End date
    Selects the ending date for the range values of your partitions. If you selected Left boundary on the Map Partitions page, this date will be the last value for each filegroup/partition. If you selected Right boundary on the Map Partitions page, this date will be the first value in the next-to-last filegroup.

    Date range
    Selects the date granularity or range value increment you want for each partition.

    After completing this page, select Next.

  8. In the Select an Output Option page, specify how you want to complete your partitioned table. Select Create Script to create a SQL script based the previous pages in the wizard. Select Run immediately to create the new partitioned table after completing all remaining pages in the wizard. Select Schedule to create the new partitioned table at a predetermined time in the future.

    If you select Create script, the following options are available under Script options:

    Script to file
    Generates the script as a .sql file. Enter a file name and location in the File name box or select Browse to open the Script File Location dialog box. From Save As, select Unicode text or ANSI text.

    Script to Clipboard
    Saves the script to the Clipboard.

    Script to New Query Window
    Generates the script to a new Query Editor window. This is the default selection.

    If you select Schedule, select Change schedule.

    1. In the New Job Schedule dialog box, in the Name box, enter the job schedule's name.

    2. On the Schedule type list, select the type of schedule:

      • Start automatically when SQL Server Agent starts

      • Start whenever the CPUs become idle

      • Recurring. Select this option if your new partitioned table updates with new information regularly.

      • One time. This is the default selection.

    3. Select or clear the Enabled check box to enable or disable the schedule.

    4. If you select Recurring:

      1. Under Frequency, on the Occurs list, specify the frequency of occurrence:

        • If you select Daily, in the Recurs every box, enter how often the job schedule repeats in days.

        • If you select Weekly, in the Recurs every box, enter how often the job schedule repeats in weeks. Select the day or days of the week on which the job schedule is run.

        • If you select Monthly, select either Day or The.

          • If you select Day, enter both the date of the month you want the job schedule to run and how often the job schedule repeats in months. For example, if you want the job schedule to run on the 15th day of the month every other month, select Day and enter "15" in the first box and "2" in the second box. The largest number allowed in the second box is "99."

          • If you select The, select the specific day of the week within the month that you want the job schedule to run and how often the job schedule repeats in months. For example, if you want the job schedule to run on the last weekday of the month every other month, select Day, select last from the first list and weekday from the second list, and then enter "2" in the last box. You can also select first, second, third, or fourth, as well as specific weekdays (for example: Sunday or Wednesday) from the first two lists. The largest number allowed in the last box is "99."

      2. Under Daily frequency, specify how often the job schedule repeats on the day the job schedule runs:

        • If you select Occurs once at, enter the specific time of day when the job schedule should run in the Occurs once at box. Enter the hour, minute, and second of the day, as well as AM or PM.

        • If you select Occurs every, specify how often the job schedule runs during the day chosen under Frequency. For example, if you want the job schedule to repeat every 2 hours during the day that the job schedule is run, select Occurs every, enter "2" in the first box, and then select hour(s) from the list. From this list you can also select minute(s) and second(s). The largest number allowed in the first box is "100."

          In the Starting at box, enter the time that the job schedule should start running. In the Ending at box, enter the time that the job schedule should stop repeating. Enter the hour, minute, and second of the day, as well as AM or PM.

      3. Under Duration, in Start date, enter the date that you want the job schedule to start running. Select End date or No end date to indicate when the job schedule should stop running. If you select End date, enter the date that you want to job schedule to stop running.

    5. If you select One Time, under One-time occurrence, in the Date box, enter the date that the job schedule will be run. In the Time box, enter the time that the job schedule will be run. Enter the hour, minute, and second of the day, as well as AM or PM.

    6. Under Summary, in Description, verify that all job schedule settings are correct.

    7. Select OK.

    After completing this page, select Next.

  9. On the Review Summary page, under Review your selections, expand all available options to verify that all partition settings are correct. If everything is as expected, select Finish.

  10. On the Create Partition Wizard Progress page, monitor status information about the actions of the Create Partition Wizard. Depending on the options that you selected in the wizard, the progress page might contain one or more actions. The top box displays the overall status of the wizard and the number of status, error, and warning messages that the wizard has received.

    The following options are available on the Create Partition Wizard Progress page:

    Details
    Provides the action, status, and any messages that are returned from action taken by the wizard.

    Action
    Specifies the type and name of each action.

    Status
    Indicates whether the wizard action as a whole returned the value of Success or Failure.

    Message
    Provides any error or warning messages that are returned from the process.

    Report
    Creates a report that contains the results of the Create Partition Wizard. The options are View Report, Save Report to File, Copy Report to Clipboard, and Send Report as Email.

    View Report
    Opens the View Report dialog box, which contains a text report of the progress of the Create Partition Wizard.

    Save Report to File
    Opens the Save Report As dialog box.

    Copy Report to Clipboard
    Copies the results of the wizard's progress report to the Clipboard.

    Send Report as Email
    Copies the results of the wizard's progress report into an email message.

    When complete, select Close.

The Create Partition Wizard creates the partition function and scheme and then applies the partitioning to the specified table. To verify the table partitioning, in Object Explorer, right-click the table and select Properties. Select the Storage page. The page displays information such as the name of the partition function and scheme and the number of partitions.

Query metadata of partitioned tables and indexes

You can query metadata to determine if a table is partitioned, the boundary points for a partitioned table, the partitioning column for a partitioned table, the number of rows in each partition, and if data compression has been implemented on partitions.

Determine if a table is partitioned

The following query returns one or more rows if the table PartitionTable is partitioned, or if any nonclustered indexes on the table are partitioned. If the table isn't partitioned and no nonclustered indexes on the table are partitioned, no rows are returned.

SELECT SCHEMA_NAME(t.schema_id) AS SchemaName, *
FROM sys.tables AS t
JOIN sys.indexes AS i
    ON t.[object_id] = i.[object_id]
JOIN sys.partition_schemes ps
    ON i.data_space_id = ps.data_space_id
WHERE t.name = 'PartitionTable';
GO

Determine the boundary values for a partitioned table

The following query returns the boundary values for each partition in the PartitionTable table.

The query uses the type column in sys.indexes to return only information for the clustered index of the table, or for the base table if the table is a heap. To include any partitioned nonclustered indexes in the query results, remove or comment out AND i.type <= 1 from the query.

SELECT SCHEMA_NAME(t.schema_id) AS SchemaName, t.name AS TableName, i.name AS IndexName,
    p.partition_number, p.partition_id, i.data_space_id, f.function_id, f.type_desc,
    r.boundary_id, r.value AS BoundaryValue
FROM sys.tables AS t
JOIN sys.indexes AS i
    ON t.object_id = i.object_id
JOIN sys.partitions AS p
    ON i.object_id = p.object_id AND i.index_id = p.index_id
JOIN  sys.partition_schemes AS s
    ON i.data_space_id = s.data_space_id
JOIN sys.partition_functions AS f
    ON s.function_id = f.function_id
LEFT JOIN sys.partition_range_values AS r
    ON f.function_id = r.function_id and r.boundary_id = p.partition_number
WHERE
    t.name = 'PartitionTable'
    AND i.type <= 1
ORDER BY SchemaName, t.name, i.name, p.partition_number;

Determine the partition column for a partitioned table

The following query returns the name of the partitioning column for table PartitionTable.

The query uses the type column in sys.indexes to return only information for the clustered index of the table, or for the base table if the table is a heap. To include any partitioned nonclustered indexes in the query results, remove or comment out AND i.type <= 1 from the query.

SELECT
    t.[object_id] AS ObjectID
    , SCHEMA_NAME(t.schema_id) AS SchemaName
    , t.name AS TableName
    , ic.column_id AS PartitioningColumnID
    , c.name AS PartitioningColumnName
    , i.name as IndexName
FROM sys.tables AS t
JOIN sys.indexes AS i
    ON t.[object_id] = i.[object_id]
    AND i.[type] <= 1 -- clustered index or a heap
JOIN sys.partition_schemes AS ps
    ON ps.data_space_id = i.data_space_id
JOIN sys.index_columns AS ic
    ON ic.[object_id] = i.[object_id]
    AND ic.index_id = i.index_id
    AND ic.partition_ordinal >= 1 -- because 0 = non-partitioning column
JOIN sys.columns AS c
    ON t.[object_id] = c.[object_id]
    AND ic.column_id = c.column_id
WHERE t.name = 'PartitionTable';
GO

Determine the rows describe the possible range of values in each partition

The following query returns the rows by partition for table PartitionTable, and a description of the 'comparison operators' for the partition function in use. Query original provided by Kalen Delaney.

The query uses the type column in sys.indexes to return only information for the clustered index of the table, or for the base table if the table is a heap. To include any partitioned nonclustered indexes in the query results, remove or comment out AND i.type <= 1 from the query.

SELECT SCHEMA_NAME(t.schema_id) AS SchemaName, t.name AS TableName, i.name AS IndexName,
    p.partition_number AS PartitionNumber, f.name AS PartitionFunctionName, p.rows AS Rows, rv.value AS BoundaryValue,
CASE WHEN ISNULL(rv.value, rv2.value) IS NULL THEN 'N/A'
ELSE
    CASE WHEN f.boundary_value_on_right = 0 AND rv2.value IS NULL THEN '>='
        WHEN f.boundary_value_on_right = 0 THEN '>'
        ELSE '>='
    END + ' ' + ISNULL(CONVERT(varchar(64), rv2.value), 'Min Value') + ' ' +
        CASE f.boundary_value_on_right WHEN 1 THEN 'and <'
                ELSE 'and <=' END
        + ' ' + ISNULL(CONVERT(varchar(64), rv.value), 'Max Value')
END AS TextComparison
FROM sys.tables AS t
JOIN sys.indexes AS i
    ON t.object_id = i.object_id
JOIN sys.partitions AS p
    ON i.object_id = p.object_id AND i.index_id = p.index_id
JOIN  sys.partition_schemes AS s
    ON i.data_space_id = s.data_space_id
JOIN sys.partition_functions AS f
    ON s.function_id = f.function_id
LEFT JOIN sys.partition_range_values AS r
    ON f.function_id = r.function_id and r.boundary_id = p.partition_number
LEFT JOIN sys.partition_range_values AS rv
    ON f.function_id = rv.function_id
    AND p.partition_number = rv.boundary_id
LEFT JOIN sys.partition_range_values AS rv2
    ON f.function_id = rv2.function_id
    AND p.partition_number - 1= rv2.boundary_id
WHERE
    t.name = 'PartitionTable'
    AND i.type <= 1
ORDER BY t.name, p.partition_number;

The TextComparison column describes the possible range of values in each partition based on the definition of the partition function. Here's a view of example results from the query:

SchemaName TableName IndexName PartitionNumber PartitionFunctionName rows BoundaryValue TextComparison
dbo PartitionTable PK_PartitionTable 1 PFTest 0 2022-03-01 00:00:00.000 >= Min Value and < Mar 1 2022 12:00AM
dbo PartitionTable PK_PartitionTable 2 PFTest 2 2022-04-01 00:00:00.000 >= Mar 1 2022 12:00AM and < Apr 1 2022 12:00AM
dbo PartitionTable PK_PartitionTable 3 PFTest 1 2022-05-01 00:00:00.000 >= Apr 1 2022 12:00AM and < May 1 2022 12:00AM
dbo PartitionTable PK_PartitionTable 4 PFTest 0 2022-06-01 00:00:00.000 >= May 1 2022 12:00AM and < Jun 1 2022 12:00AM
dbo PartitionTable PK_PartitionTable 5 PFTest 1 2022-07-01 00:00:00.000 >= Jun 1 2022 12:00AM and < Jul 1 2022 12:00AM
dbo PartitionTable PK_PartitionTable 6 PFTest 0 NULL >= Jul 1 2022 12:00AM and < Max Value

Limitations

Learn about limitations as well as performance considerations for partitioning in Limitations