You must specify at least one range partition using the PARTITION clause. The range partitioning key value determines the high value of the range partitions, which is called the transition point, and the database automatically creates interval partitions for data beyond that transition point. The lower boundary of every interval partition is the non-inclusive upper boundary of the previous range or interval partition.

For example, if you create an interval partitioned table with monthly intervals and the transition point is at January 1, 2010, then the lower boundary for the January 2010 interval is January 1, 2010. The lower boundary for the July 2010 interval is July 1, 2010, regardless of whether the June 2010 partition was previously created. Note, however, that using a date where the high or low bound of the partition would be out of the range set for storage causes an error. For example, TO_DATE('9999-12-01', 'YYYY-MM-DD') causes the high bound to be 10000-01-01, which would not be storable if 10000 is out of the legal range.

The optional STORE IN clause lets you specify one or more tablespaces into which the database stores interval partition data using a round-robin algorithm for subsequently created interval partitions.

For interval partitioning, you can specify only one partitioning key column and the datatype is restricted.

The following example specifies four partitions with varying interval widths. It also specifies that above the transition point of January 1, 2010, partitions are created with an interval width of one month. The high bound of partition p3 represents the transition point. p3 and all partitions below it (p0, p1, and p2 in this example) are in the range section while all partitions above it fall into the interval section.

CREATE TABLE interval_sales
    ( prod_id        NUMBER(6)
    , cust_id        NUMBER
    , time_id        DATE
    , channel_id     CHAR(1)
    , promo_id       NUMBER(6)
    , quantity_sold  NUMBER(3)
    , amount_sold    NUMBER(10,2)
    ) 
  PARTITION BY RANGE (time_id) 
  INTERVAL(NUMTOYMINTERVAL(1, 'MONTH'))
    ( PARTITION p0 VALUES LESS THAN (TO_DATE('1-1-2008', 'DD-MM-YYYY')),
      PARTITION p1 VALUES LESS THAN (TO_DATE('1-1-2009', 'DD-MM-YYYY')),
      PARTITION p2 VALUES LESS THAN (TO_DATE('1-7-2009', 'DD-MM-YYYY')),
      PARTITION p3 VALUES LESS THAN (TO_DATE('1-1-2010', 'DD-MM-YYYY')) );

The PARTITION BY REFERENCE clause specifies the name of a referential constraint and this constraint becomes the partitioning referential constraint that is used as the basis for reference partitioning in the table. The referential constraint must be enabled and enforced.

As with other partitioned tables, you can specify object-level default attributes, and you can optionally specify partition descriptors that override the object-level defaults on a per-partition basis.

Example 4-9 creates a parent table orders which is range-partitioned on order_date. The reference-partitioned child table order_items is created with four partitions, Q1_2005, Q2_2005, Q3_2005, and Q4_2005, where each partition contains the order_items rows corresponding to orders in the respective parent partition.

If partition descriptors are provided, then the number of partitions described must exactly equal the number of partitions or subpartitions in the referenced table. If the parent table is a composite partitioned table, then the table has one partition for each subpartition of its parent; otherwise the table has one partition for each partition of its parent.

Partition bounds cannot be specified for the partitions of a reference-partitioned table.

The partitions of a reference-partitioned table can be named. If a partition is not explicitly named, then it inherits its name from the corresponding partition in the parent table, unless this inherited name conflicts with an existing explicit name. In this case, the partition has a system-generated name.

Partitions of a reference-partitioned table collocate with the corresponding partition of the parent table, if no explicit tablespace is specified for the reference-partitioned table's partition.

When creating an interval partition in a child table, the partition name is inherited from the associated parent table fragment. If the child table has a table-level default tablespace, then it is used as tablespace for the new interval partition; otherwise, the tablespace is inherited from the parent table fragment.

The SQL ALTER TABLE SET INTERVAL statement is not allowed for reference-partitioned tables, but can be run on tables that have reference-partitioned children. In particular, ALTER TABLE SET INTERVAL removes the interval property from the targeted table and converts any interval-reference children to ordinary reference-partitioned tables. Also, the SQL ALTER TABLE SET STORE IN statement is not allowed for reference-partitioned tables, but can be run on tables that have reference-partitioned children.

Operations that transform interval partitions to conventional partitions in the parent table, such as ALTER TABLE SPLIT PARTITION on an interval partition, construct the corresponding transformation in the child table, creating partitions in the child table as necessary.

For example, the following SQL statements provides three interval partitions in the parent table and none in the child table:

CREATE TABLE par(pk INT CONSTRAINT par_pk PRIMARY KEY, i INT)
 PARTITION BY RANGE(i) INTERVAL (10)
 (PARTITION p1 VALUES LESS THAN (10));

CREATE TABLE chi(fk INT NOT NULL, i INT,
 CONSTRAINT chi_fk FOREIGN KEY(fk) REFERENCES par(pk))
 PARTITION BY REFERENCE(chi_fk);

INSERT INTO par VALUES(15, 15);
INSERT INTO par VALUES(25, 25);
INSERT INTO par VALUES(35, 35);

You can display information about partitions with the USER_TAB_PARTITIONS view:

SELECT table_name, partition_position, high_value, interval
   FROM USER_TAB_PARTITIONS WHERE table_name IN ('PAR', 'CHI')
   ORDER BY 1, 2;

TABLE_NAME       PARTITION_POSITION HIGH_VALUE INT
---------------- ------------------ ---------- ---
CHI                               1            NO
PAR                               1         10 NO
PAR                               2         20 YES
PAR                               3         30 YES
PAR                               4         40 YES

If the interval partition is split in the parent table, then some interval partitions are converted to conventional partitions for all tables in the hierarchy, creating conventional partitions in the child table in the process. For example:

ALTER TABLE par SPLIT PARTITION FOR (25) AT (25)
   INTO (partition x, partition y);

SELECT table_name, partition_position, high_value, interval
   FROM USER_TAB_PARTITIONS WHERE table_name IN ('PAR', 'CHI')
   ORDER BY 1, 2;

TABLE_NAME       PARTITION_POSITION HIGH_VALUE INT
---------------- ------------------ ---------- ---
CHI                               1            NO
CHI                               2            NO
CHI                               3            NO
CHI                               4            NO
PAR                               1         10 NO
PAR                               2         20 NO
PAR                               3         25 NO
PAR                               4         30 NO
PAR                               5         40 YES

Interval-reference functionality requires that the database compatibility level (Oracle Database COMPATIBLE initialization parameter setting) be set to greater than or equal to 12.0.0.0.

The following example specifies that individual partitions are loaded into the In-Memory Column Store using the INMEMORY clause with the partitioning clauses of the CREATE TABLE SQL statements.

CREATE TABLE list_customers 
   ( customer_id             NUMBER(6)
   , cust_first_name         VARCHAR2(20) 
   , cust_last_name          VARCHAR2(20)
   , cust_address            CUST_ADDRESS_TYP
   , nls_territory           VARCHAR2(30)
   , cust_email              VARCHAR2(40))
   PARTITION BY LIST (nls_territory) (
   PARTITION asia VALUES ('CHINA', 'THAILAND') 
         INMEMORY MEMCOMPRESS FOR CAPACITY HIGH,
   PARTITION europe VALUES ('GERMANY', 'ITALY', 'SWITZERLAND') 
         INMEMORY MEMCOMPRESS FOR CAPACITY LOW,
   PARTITION west VALUES ('AMERICA') 
         INMEMORY MEMCOMPRESS FOR CAPACITY LOW,
   PARTITION east VALUES ('INDIA') 
         INMEMORY MEMCOMPRESS FOR CAPACITY HIGH,
   PARTITION rest VALUES (DEFAULT);

All partition methods are supported when using virtual columns, including interval partitioning and all different combinations of composite partitioning. A virtual column used as the partitioning column cannot use calls to a PL/SQL function.

The following example shows the sales table partitioned by range-range using a virtual column for the subpartitioning key. The virtual column calculates the total value of a sale by multiplying amount_sold and quantity_sold. As the example shows, row movement is also supported with virtual columns. If row movement is enabled, then a row migrates from one partition to another partition if the virtual column evaluates to a value that belongs to another partition.

CREATE TABLE sales
  ( prod_id       NUMBER(6) NOT NULL
  , cust_id       NUMBER NOT NULL
  , time_id       DATE NOT NULL
  , channel_id    CHAR(1) NOT NULL
  , promo_id      NUMBER(6) NOT NULL
  , quantity_sold NUMBER(3) NOT NULL
  , amount_sold   NUMBER(10,2) NOT NULL
  , total_amount AS (quantity_sold * amount_sold)
  )
 PARTITION BY RANGE (time_id) INTERVAL (NUMTOYMINTERVAL(1,'MONTH'))
 SUBPARTITION BY RANGE(total_amount)
 SUBPARTITION TEMPLATE
   ( SUBPARTITION p_small VALUES LESS THAN (1000)
   , SUBPARTITION p_medium VALUES LESS THAN (5000)
   , SUBPARTITION p_large VALUES LESS THAN (10000)
   , SUBPARTITION p_extreme VALUES LESS THAN (MAXVALUE)
   )
 (PARTITION sales_before_2007 VALUES LESS THAN
        (TO_DATE('01-JAN-2007','dd-MON-yyyy'))
)
ENABLE ROW MOVEMENT
PARALLEL NOLOGGING;

The compression attribute can be declared for a tablespace, a table, or a partition of a table. Whenever the compress attribute is not specified, it is inherited like any other storage attribute.

Example 4-13 creates a range-partitioned table with one compressed partition costs_old. The compression attribute for the table and all other partitions is inherited from the tablespace level.

Key compression is applicable only to B-tree indexes. Bitmap indexes are stored in a compressed manner by default. An index using key compression eliminates repeated occurrences of key column prefix values, thus saving space and I/O.

The following example creates a local partitioned index with all partitions except the most recent one compressed:

CREATE INDEX i_cost1 ON costs_demo (prod_id) COMPRESS LOCAL
   (PARTITION costs_old, PARTITION costs_q1_2003, 
    PARTITION costs_q2_2003, PARTITION costs_recent NOCOMPRESS);

You cannot specify COMPRESS (or NOCOMPRESS) explicitly for an index subpartition. All index subpartitions of a given partition inherit the key compression setting from the parent partition.

To modify the key compression attribute for all subpartitions of a given partition, you must first issue an ALTER INDEX...MODIFY PARTITION statement and then rebuild all subpartitions. The MODIFY PARTITION clause marks all index subpartitions as UNUSABLE.

For the purposes of this discussion, the term Collection Tables is used for the following two categories: (1) ordered collection tables inside XMLType tables or columns, and (2) nested tables inside object tables or columns.

When you partition Collection Tables, Oracle Database uses the partitioning scheme of the base table. Also, Collection Tables are automatically partitioned when the base table is partitioned. DML against a partitioned nested table behaves in a similar manner to that of a reference partitioned table.

Oracle Database provides a LOCAL keyword to equipartition a Collection Table with a partitioned base table. This is the default behavior in this release. The default in earlier releases was not to equipartition the Collection Table with the partitioned base table. Now you must specify the GLOBAL keyword to store an unpartitioned Collection Table with a partitioned base table.

Out-of-line (OOL) table partitioning is supported. However, you cannot create two tables of the same XML schema that has out-of-line tables. This restriction means that exchange partitioning cannot be performed for schemas with OOL tables because it is not possible to have two tables of the same schema.

The statement in the following example creates a nested table partition.

CREATE TABLE print_media_part (
   product_id NUMBER(6),
   ad_id NUMBER(6),
   ad_composite BLOB,
   ad_sourcetext CLOB,
   ad_finaltext CLOB,
   ad_fltextn NCLOB,
   ad_textdocs_ntab TEXTDOC_TAB,
   ad_photo BLOB,
   ad_graphic BFILE,
   ad_header ADHEADER_TYP)
NESTED TABLE ad_textdocs_ntab STORE AS textdoc_nt
PARTITION BY RANGE (product_id)
  (PARTITION p1 VALUES LESS THAN (100),
   PARTITION p2 VALUES LESS THAN (200));