2 Partitioning Concepts

Basics of Partitioning

From the perspective of a database administrator, a partitioned object has multiple pieces that can be managed either collectively or individually. This gives an administrator considerable flexibility in managing partitioned objects. However, from the perspective of the application, a partitioned table is identical to a nonpartitioned table; no modifications are necessary when accessing a partitioned table using SQL queries and DML statements.

Figure 2-1 offers a graphical view of how partitioned tables differ from nonpartitioned tables.

Figure 2-1 A View of Partitioned Tables

Description of "Figure 2-1 A View of Partitioned Tables"

Partitioning Key

Each row in a partitioned table is unambiguously assigned to a single partition. The partitioning key consists of one or more columns that determine the partition where each row is stored. Oracle automatically directs insert, update, and delete operations to the appropriate partition with the partitioning key.

Partitioned Tables

Any table can be partitioned into a million separate partitions except those tables containing columns with LONG or LONG RAW data types. You can, however, use tables containing columns with CLOB or BLOB data types.

This sections contains the following topics:

• When to Partition a Table

• When to Partition an Index

Partitioned Index-Organized Tables

Partitioned index-organized tables are very useful for providing improved performance, manageability, and availability for index-organized tables.

For partitioning an index-organized table:

• Partition columns must be a subset of the primary key columns.

• Secondary indexes can be partitioned (both locally and globally).

• OVERFLOW data segments are always equipartitioned with the table partitions.

System Partitioning

System partitioning enables application-controlled partitioning without having the database controlling the data placement. The database simply provides the ability to break down a table into partitions without knowing what the individual partitions are going to be used for. All aspects of partitioning have to be controlled by the application. For example, an attempt to insert into a system partitioned table without the explicit specification of a partition fails.

System partitioning provides the well-known benefits of partitioning (scalability, availability, and manageability), but the partitioning and actual data placement are controlled by the application.

Partitioning for Information Lifecycle Management

Information Lifecycle Management (ILM) is concerned with managing data during its lifetime. Partitioning plays a key role in ILM because it enables groups of data (that is, partitions) to be distributed across different types of storage devices and managed individually.

For more information about Information Lifecycle Management, see Chapter 5, "Using Partitioning for Information Lifecycle Management".

Partitioning and LOB Data

Unstructured data (such as images and documents) which is stored in a LOB column in a database can also be partitioned. When a table is partitioned, all of the columns reside in the tablespace for that partition, except LOB columns, which can be stored in their own tablespace.

This technique is very useful when a table consists of large LOBs because they can be stored separately from the main data. This can be beneficial if the main data is being frequently updated but the LOB data is not. For example, an employee record may contain a photo which is unlikely to change frequently. However, the employee personnel details (such as address, department, manager, and so on) could change. This approach also means that less expensive storage can be used for storing the LOB data and more expensive, faster storage can be used for the employee record.

Collections in XMLType and Object Data

Partitioning when using XMLType and object tables and columns offers the standard advantages of partitioning, such as enabling tables and indexes to be subdivided into smaller pieces, thus enabling these database objects to be managed and accessed at a finer level of granularity.

When you partition an XMLType table or a table with an XMLType column using list, range, or hash partitioning, any ordered collection tables (OCTs) within the data are automatically partitioned accordingly, by default. This equipartitioning means that the partitioning of an OCT follows the partitioning scheme of its parent (base) table. There is a corresponding collection-table partition for each partition of the base table. A child element is stored in the collection-table partition that corresponds to the base-table partition of its parent element.

If you partition a table that has a nested table, then Oracle Database uses the partitioning scheme of the original base table as the basis for how the nested table is partitioned. This partitioning of one base table partition for each nested table partition is called equipartitioning. By default, nested tables are automatically partitioned when the base table is partitioned. Note, however, that composite partitioning is not supported for OCTs or nested tables.

For information about partitioning an XMLType table, refer to "Partitioning of Collections in XMLType and Objects".

Partitioning for Performance

By limiting the amount of data to be examined or operated on, and by providing data distribution for parallel execution, partitioning provides multiple performance benefits. Partitioning features include:

• Partition Pruning

• Partition-Wise Joins

Partitioning for Manageability

Partitioning enables you to partition tables and indexes into smaller, more manageable units, providing database administrators with the ability to pursue a divide and conquer approach to data management. With partitioning, maintenance operations can be focused on particular portions of tables. For example, you could back up a single partition of a table, rather than back up the entire table. For maintenance operations across an entire database object, it is possible to perform these operations on a per-partition basis, thus dividing the maintenance process into more manageable chunks.

A typical usage of partitioning for manageability is to support a rolling window load process in a data warehouse. Suppose that you load new data into a table on a weekly basis. That table could be partitioned so that each partition contains one week of data. The load process is simply the addition of a new partition using a partition exchange load. Adding a single partition is much more efficient than modifying the entire table, because you do not need to modify any other partitions.

Partitioning for Availability

Partitioned database objects provide partition independence. This characteristic of partition independence can be an important part of a high-availability strategy. For example, if one partition of a partitioned table is unavailable, then all of the other partitions of the table remain online and available. The application can continue to execute queries and transactions against the available partitions for the table, and these database operations can run successfully, provided they do not need to access the unavailable partition.

The database administrator can specify that each partition be stored in a separate tablespace; the most common scenario is having these tablespaces stored on different storage tiers. Storing different partitions in different tablespaces enables you to do backup and recovery operations on each individual partition, independent of the other partitions in the table. Thus allowing the active parts of the database to be made available sooner so access to the system can continue, while the inactive data is still being restored. Moreover, partitioning can reduce scheduled downtime. The performance gains provided by partitioning may enable you to complete maintenance operations on large database objects in relatively small batch windows.

Single-Level Partitioning

A table is defined by specifying one of the following data distribution methodologies, using one or more columns as the partitioning key:

• Range Partitioning

• Hash Partitioning

• List Partitioning

For example, consider a table with a column of type NUMBER as the partitioning key and two partitions less_than_five_hundred and less_than_one_thousand. The less_than_one_thousand partition contains rows where the following condition is true:

500 <= partitioning key < 1000

Figure 2-2 offers a graphical view of the basic partitioning strategies for a single-level partitioned table.

Figure 2-2 List, Range, and Hash Partitioning

Description of "Figure 2-2 List, Range, and Hash Partitioning"

Composite Partitioning

Composite partitioning is a combination of the basic data distribution methods; a table is partitioned by one data distribution method and then each partition is further subdivided into subpartitions using a second data distribution method. All subpartitions for a given partition represent a logical subset of the data.

Composite partitioning supports historical operations, such as adding new range partitions, but also provides higher degrees of potential partition pruning and finer granularity of data placement through subpartitioning. Figure 2-3 offers a graphical view of range-hash and range-list composite partitioning, as an example.

Figure 2-3 Composite Partitioning

Description of "Figure 2-3 Composite Partitioning"

This section describes the following types of partitioning:

• Composite Range-Range Partitioning

• Composite Range-Hash Partitioning

• Composite Range-List Partitioning

• Composite List-Range Partitioning

• Composite List-Hash Partitioning

• Composite List-List Partitioning

Manageability Extensions

The following extensions significantly enhance the manageability of partitioned tables:

• Interval Partitioning

• Partition Advisor

Partitioning Key Extensions

The following extensions extend the flexibility in defining partitioning keys:

• Reference Partitioning

• Virtual Column-Based Partitioning

Reference Partitioning

Reference partitioning enables the partitioning of two tables that are related to one another by referential constraints. The partitioning key is resolved through an existing parent-child relationship, enforced by enabled and active primary key and foreign key constraints.

The benefit of this extension is that tables with a parent-child relationship can be logically equipartitioned by inheriting the partitioning key from the parent table without duplicating the key columns. The logical dependency also automatically cascades partition maintenance operations, thus making application development easier and less error-prone.

An example of reference partitioning is the Orders and LineItems tables related to each other by a referential constraint orderid_refconstraint. Namely, LineItems.order_id references Orders.order_id. The Orders table is range partitioned on order_date. Reference partitioning on orderid_refconstraint for LineItems leads to creation of the following partitioned table, which is equipartitioned on the Orders table, as shown in Figure 2-4 and Figure 2-5.

Figure 2-4 Before Reference Partitioning

Description of "Figure 2-4 Before Reference Partitioning"

Figure 2-5 With Reference Partitioning

Description of "Figure 2-5 With Reference Partitioning"

All basic partitioning strategies are available for reference partitioning. Interval partitioning cannot be used with reference partitioning.

Deciding on the Type of Partitioned Index to Use

When deciding what kind of partitioned index to use, you should consider the following guidelines in this order:

1. If the table partitioning column is a subset of the index keys, then use a local index. If this is the case, then you are finished. If this is not the case, then continue to guideline 2.

2. If the index is unique and does not include the partitioning key columns, then use a global index. If this is the case, then you are finished. Otherwise, continue to guideline 3.

3. If your priority is manageability, then use a local index. If this is the case, then you are finished. If this is not the case, continue to guideline 4.

4. If the application is an OLTP type and users need quick response times, then use a global index. If the application is a DSS type and users are more interested in throughput, then use a local index.

For more information about partitioned indexes and how to decide which type to use, refer to Chapter 6, "Using Partitioning in a Data Warehouse Environment" and Chapter 7, "Using Partitioning in an Online Transaction Processing Environment".

Local Partitioned Indexes

Local partitioned indexes are easier to manage than other types of partitioned indexes. They also offer greater availability and are common in DSS environments. The reason for this is equipartitioning: each partition of a local index is associated with exactly one partition of the table. This functionality enables Oracle to automatically keep the index partitions synchronized with the table partitions, and makes each table-index pair independent. Any actions that make one partition's data invalid or unavailable only affect a single partition.

Local partitioned indexes support more availability when there are partition or subpartition maintenance operations on the table. A type of index called a local nonprefixed index is very useful for historical databases. In this type of index, the partitioning is not on the left prefix of the index columns. For more information about prefixed indexes, refer to "Index Partitioning".

You cannot explicitly add a partition to a local index. Instead, new partitions are added to local indexes only when you add a partition to the underlying table. Likewise, you cannot explicitly drop a partition from a local index. Instead, local index partitions are dropped only when you drop a partition from the underlying table.

A local index can be unique. However, in order for a local index to be unique, the partitioning key of the table must be part of the index's key columns.

Figure 2-6 offers a graphical view of local partitioned indexes.

Figure 2-6 Local Partitioned Index

Description of "Figure 2-6 Local Partitioned Index"

For more information about local partitioned indexes, refer to "Local Partitioned Indexes".

Global Partitioned Indexes

Oracle offers global range partitioned indexes and global hash partitioned indexes, discussed in the following topics:

• Global Range Partitioned Indexes

• Global Hash Partitioned Indexes

• Maintenance of Global Partitioned Indexes

Maintenance of Global Partitioned Indexes

By default, the following operations on partitions on a heap-organized table mark all global indexes as unusable:

ADD (HASH) 
COALESCE (HASH) 
DROP 
EXCHANGE 
MERGE 
MOVE 
SPLIT 
TRUNCATE 

These indexes can be maintained by appending the clause UPDATE INDEXES to the SQL statements for the operation. The two advantages to maintaining global indexes:

• The index remains available and online throughout the operation. Hence no other applications are affected by this operation.

• The index does not have to be rebuilt after the operation.

Note:

This feature is supported only for heap-organized tables.

Figure 2-7 offers a graphical view of global partitioned indexes.

Figure 2-7 Global Partitioned Index

Description of "Figure 2-7 Global Partitioned Index"

For more information about global partitioned indexes, refer to "Global Partitioned Indexes".

Global Nonpartitioned Indexes

Global nonpartitioned indexes behave just like a nonpartitioned index.

Figure 2-8 offers a graphical view of global nonpartitioned indexes.

Figure 2-8 Global Nonpartitioned Index

Description of "Figure 2-8 Global Nonpartitioned Index"

Miscellaneous Information about Creating Indexes on Partitioned Tables

You can create bitmap indexes on partitioned tables, with the restriction that the bitmap indexes must be local to the partitioned table. They cannot be global indexes.

Global indexes can be unique. Local indexes can only be unique if the partitioning key is a part of the index key.

Partitioned Indexes on Composite Partitions

Here are a few points to remember when using partitioned indexes on composite partitions:

• Subpartitioned indexes are always local and stored with the table subpartition by default.

• Tablespaces can be specified at either index or index subpartition levels.