5 Indexing and Querying Spatial Data
After you have loaded spatial data , you should create a spatial index on it to enable efficient query performance using the data.
Topics:
- Creating a Spatial Index
Once data has been loaded into the spatial tables through either bulk or transactional loading, a spatial index (that is, a spatial R-tree index) must be created on each geometry column in the tables for efficient access to the data. - Querying Spatial Data
The structures of a spatial layer are used to resolve spatial queries and spatial joins.
Related Topics
5.1 Creating a Spatial Index
Once data has been loaded into the spatial tables through either bulk or transactional loading, a spatial index (that is, a spatial R-tree index) must be created on each geometry column in the tables for efficient access to the data.
For example, the following statement creates a spatial index named territory_idx
using default values for all parameters:
CREATE INDEX territory_idx ON territories (territory_geom) INDEXTYPE IS MDSYS.SPATIAL_INDEX_V2;
Note:
For an explanation of the “_V2” in INDEXTYPE IS MDSYS.SPATIAL_INDEX_V2, see Using System-Managed Spatial Indexes
For detailed information about options for creating a spatial index, see the documentation for the CREATE INDEX statement.
If the index creation does not complete for any reason, the index is invalid and must be deleted with the DROP INDEX <index_name> [FORCE] statement.
Within each geometry column to be indexed, all the geometries must have the same SDO_SRID value.
Spatial indexes can be built on two, three, or four dimensions of data. The default number of dimensions is two, but if the data has more than two dimensions, you can use the sdo_indx_dims
parameter keyword to specify the number of dimensions on which to build the index. (For information about support for three-dimensional geometries, see Three-Dimensional Spatial Objects. For an explanation of support for various combinations of dimensionality in query elements, see Data and Index Dimensionality, and Spatial Queries.)
If you are not using the automatic undo management feature or the PGA memory management feature, or both, of Oracle Database, see Rollback Segments and Sort Area Size for information about initialization parameter values that you may need to set. Both automatic undo management and PGA memory management are enabled by default, and their use is highly recommended.
The tablespace specified with the tablespace
keyword in the CREATE INDEX statement (or the default tablespace if the tablespace
keyword is not specified) is used to hold both the index data table and some transient tables that are created for internal computations. If you specify WORK_TABLESPACE as the tablespace, the transient tables are stored in the work tablespace.
For large tables (over 1 million rows), a temporary tablespace may be needed to perform internal sorting operations. The recommended size for this temporary tablespace is 100*n bytes, where n is the number of rows in the table, up to a maximum requirement of 1 gigabyte of temporary tablespace.
To estimate the space that will be needed to create a spatial index, use the SDO_TUNE.ESTIMATE_RTREE_INDEX_SIZE function.
This section also contains the following topics.
Topics:
- Using System-Managed Spatial Indexes
- Constraining Data to a Geometry Type
- Creating a Composite B-tree Spatial Index on Points
- Creating a Cross-Schema Index
- Using Partitioned Spatial Indexes
- Exchanging Partitions Including Indexes
- Export and Import Considerations with Spatial Indexes and Data
- Distributed and Oracle XA Transactions Not Supported with R-Tree Spatial Indexes
- Enabling Access to Spatial Index Statistics
- Rollback Segments and Sort Area Size
5.1.1 Using System-Managed Spatial Indexes
Effective with Release 12.2, spatial indexes can be system-managed by specifying INDEXTYPE=MDSYS.SPATIAL_INDEX_V2 at index creation. You are strongly encouraged to use this index type for all new spatial indexes you create, regardless of whether the spatial table or the spatial index is partitioned, and you may also want to use it if you decide to re-create legacy spatial indexes.
The main benefit is simplified spatial index management. This is most beneficial in cases of partitioning, because this new index type eliminates the need for most, if not all, index partitioning management operations. Full support is provided for almost all Oracle Database base table partitioning models, including:
-
Single-level partitioning: range, hash, list
-
Composite partitioning: range-range, range-hash, range-list, list-range, list-hash, list-list, hash-hash, hash-list, hash-range
-
Partitioning extensions: interval (but not interval-based composite partitions), reference, virtual column-based partitioning
The old INDEXTYPE=MDSYS.SPATIAL_INDEX (without the “_V2”) is still available for use, It may provide slightly better index creation performance, especially with small data sets and no partitioning involved. You might also want to use the old type if you need to drop a legacy spatial index and then want to re-create it in exactly the same form as it was before. However, in all or almost all cases you will want to specify INDEXTYPE=MDSYS.SPATIAL_INDEX_V2 when creating any spatial index.
The following topics provide examples of using INDEXTYPE=MDSYS.SPATIAL_INDEX_V2.
Topics:
5.1.1.1 Spatial Indexing Example: Interval Partitioning
Interval partitioning is a partitioning method where Oracle Database automatically creates base table partitions when the inserted value does not match any existing partition ranges.
The following restrictions apply:
-
You can only specify one base table partitioning key column, and it must be of type NUMBER or DATE.
-
Interval partitioning is not supported for index-organized tables.
Consider the following example of a base table named DEST_TABLE, partitioned based on the month of the “currently last seen” column:
CREATE TABLE dest_table
PARTITION BY RANGE ("CURR_LAST_SEEN_AT")
INTERVAL(NUMTOYMINTERVAL(1, 'MONTH'))
(PARTITION "YEAR_1999"
VALUES LESS THAN (TIMESTAMP' 2000-01-01 00:00:00'),
PARTITION "YEAR_2000"
VALUES LESS THAN (TIMESTAMP' 2001-01-01 00:00:00'))
PARALLEL
AS SELECT imo_num,
last_seen_at curr_last_seen_at,
a.geometry.sdo_point.x curr_longitude,
a.geometry.sdo_point.y curr_latitude,
LAG(last_seen_at)
OVER (partition by imo_num ORDER BY last_seen_at) prev_last_seen_at,
LEAD(last_seen_at)
OVER (partition by imo_num ORDER BY last_seen_at) next_last_seen_at,
LAG(a.geometry.sdo_point.x)
OVER (partition by imo_num ORDER BY last_seen_at) prev_longitude,
LAG(a.geometry.sdo_point.y)
OVER (partition by imo_num ORDER BY last_seen_at) prev_latitude,
LEAD(a.geometry.sdo_point.x)
OVER (partition by imo_num ORDER BY last_seen_at) next_longitude,
LEAD(a.geometry.sdo_point.y)
OVER (partition by imo_num ORDER BY last_seen_at) next_latitude
FROM source_table a;
As data is selected from the source table (source_table) into this DEST_TABLE table, Oracle Database automatically partitions the data by the month of the CURR_LAST_SEEN_AT column. If the corresponding partition does not exist, Oracle Database will automatically create a new partition without any action required on your part.
The preceding example created two explicit partitions. To see what our actual data looks like, use a query such as for following to the database dictionary to see what partitions were created:
SQL> select partition_name, high_value
2 from user_tab_partitions
3 where table_name = 'DEST_TABLE'
4 order by partition_name;
PARTITION_NAME
--------------------------------------------------------------------------------
HIGH_VALUE
--------------------------------------------------------------------------------
SYS_P2881
TIMESTAMP' 2014-08-01 00:00:00'
SYS_P2882
TIMESTAMP' 2014-09-01 00:00:00'
SYS_P2883
TIMESTAMP' 2014-10-01 00:00:00'
SYS_P2884
TIMESTAMP' 2014-11-01 00:00:00'
YEAR_1999
TIMESTAMP' 2000-01-01 00:00:00'
YEAR_2000
TIMESTAMP' 2001-01-01 00:00:00'
6 rows selected.
Now create the spatial index. The following example uses function-based index; the function will convert the base table scalar longitude and latitude columns into a virtual spatial geometry, which will be the index “key value”:
CREATE OR REPLACE FUNCTION get_geometry(in_longitude NUMBER,
in_latitude NUMBER)
return SDO_GEOMETRY DETERMINISTIC PARALLEL_ENABLE IS
BEGIN
RETURN sdo_geometry(2001,
4326,
sdo_point_type(in_longitude, in_latitude, NULL),
NULL,
NULL);
END;
/
INSERT INTO user_sdo_geom_metadata VALUES (
'DEST_TABLE','SCOTT.GET_GEOMETRY(CURR_LONGITUDE,CURR_LATITUDE)',
SDO_DIM_ARRAY(SDO_DIM_ELEMENT('Longitude', '-180', '180', '.05'),
SDO_DIM_ELEMENT('Latitude', '-90', '90', '.05')),
4326);
COMMIT;
CREATE INDEX geom_idx1
ON dest_table(GET_GEOMETRY(CURR_LONGITUDE, CURR_LATITUDE))
INDEXTYPE IS MDSYS.SPATIAL_INDEX_V2 LOCAL PARALLEL;
Notice that no partitioning information was specified for the spatial index. To see the partitions that were automatically created, use a query such as for following:
SQL> select partition_name, high_value
2 from user_ind_partitions
3 where index_name = 'PRECOMPUTE_GEOM_IDX1'
4 order by partition_name;
PARTITION_NAME
--------------------------------------------------------------------------------
HIGH_VALUE
--------------------------------------------------------------------------------
SYS_P2921
TIMESTAMP' 2014-08-01 00:00:00'
SYS_P2922
TIMESTAMP' 2014-09-01 00:00:00'
SYS_P2923
TIMESTAMP' 2014-10-01 00:00:00'
SYS_P2924
TIMESTAMP' 2014-11-01 00:00:00'
YEAR_1999
TIMESTAMP' 2000-01-01 00:00:00'
YEAR_2000
TIMESTAMP' 2001-01-01 00:00:00'
6 rows selected.
Notice that the number of index partitions is the same as were created for the base table, including two partitions with the same name as those explicitly specified in the CREATE TABLE statement. However, the system-generated index partition names are different from the base table name.
5.1.1.2 Spatial Indexing Example: Virtual Column Partitioning
A virtual column is an expression based on one or more existing columns in the base table. While a virtual column is only stored as metadata and does not consume physical space, it can be indexed and also contain optimizer statistics and histograms. Partitioning is supported for a table using a partitioning key on a virtual column
If system-managed spatial indexing is not used, then to partition a table by using a derived value, a DBA must create and populate an additional physical column in order to achieve the same result. The derived value then must be populated by the application or by a trigger that evaluates the expression before insertion. In either case, achieving this goal without system-managed indexing requires additional overhead and increased disk space for the physical column.
If system-managed indexing is used, the ability to use an expression as a partitioning key provides a more efficient way to meet comprehensive business requirements without incurring unnecessary overhead. This can be very useful when a table cannot be partitioned by the existing data columns.
Consider the following example of a base table named ACCOUNTS that contains a virtual column named REGION:
create table accounts_v
( account_number varchar2(30),
account_name varchar2(30),
contact_person varchar2(30),
region AS (case
when substr(account_name,1,1) = 'N' then 'NORTH'
when substr(account_name,1,1) = 'E' then 'EAST'
when substr(account_name,1,1) = 'S' then 'SOUTH'
when substr(account_name,1,1) = 'W' then 'WEST'
end),
shape mdsys.sdo_geometry
)
partition by list (region)
( partition pN values ('NORTH'),
partition pE values ('EAST'),
partition pS values ('SOUTH'),
partition pW values ('WEST')
);
Now create a system-managed local domain spatial index on the SHAPE column:
insert into user_sdo_geom_metadata
values('ACCOUNTS_V',
'SHAPE',
mdsys.sdo_dim_array(
mdsys.sdo_dim_element('Longitude', -180, 180, 0.05),
mdsys.sdo_dim_element('Latitude', -90, 90, 0.05)),
NULL);
commit;
create index shape_v_idx on accounts_v(shape)
indextype is mdsys.spatial_index_v2 LOCAL;
Notice that no spatial index partition information was specified. However, a full set of spatial index partitions was created automatically and without user intervention.
To verify the placement of records in the appropriate partitions, query a specific partition. The following query is for the accounts in the East region (:
SQL> select * from accounts_v partition(PE)
2 order by account_number;
ACCOUNT_NUMBER ACCOUNT_NAME
------------------------------ ------------------------------
CONTACT_PERSON REGIO
------------------------------ -----
SHAPE(SDO_GTYPE, SDO_SRID, SDO_POINT(X, Y, Z), SDO_ELEM_INFO, SDO_ORDINATES)
--------------------------------------------------------------------------------
8778-5435-5345-5223 E-HORIZON-AUTOMOTIVE
RICK EAST
SDO_GEOMETRY(2001, NULL, SDO_POINT_TYPE(2, 2, NULL), NULL, NULL)
1 row selected.
5.1.2 Constraining Data to a Geometry Type
When you create or rebuild a spatial index, you can ensure that all geometries that are in the table or that are inserted later are of a specified geometry type. To constrain the data to a geometry type in this way, use the layer_gtype
keyword in the PARAMETERS clause of the CREATE INDEX or ALTER INDEX REBUILD statement, and specify a value from the Geometry Type column of the Valid SDO_GTYPE Values table described in SDO_GTYPE. For example, to constrain spatial data in a layer to polygons:
CREATE INDEX cola_spatial_idx ON cola_markets(shape) INDEXTYPE IS MDSYS.SPATIAL_INDEX_V2 PARAMETERS ('layer_gtype=POLYGON');
The geometry types in SDO_GTYPE are considered as a hierarchy when data is checked:
-
The MULTI forms include the regular form also. For example, specifying
'layer_gtype=
MULTIPOINT'
allows the layer to include both POINT and MULTIPOINT geometries. -
COLLECTION allows the layer to include all types of geometries.
5.1.3 Creating a Composite B-tree Spatial Index on Points
Effective with Release 12.2, you can create a composite B-tree index on point data by specifying the cbtree_index=true
and layer_gtype=POINT
keywords in the PARAMETERS clause of the CREATE INDEX statement. For example:
CREATE INDEX pt_idx on PT_CB(c2) indextype is mdsys.spatial_index_v2
PAREMETERS ('layer_gtype=POINT cbtree_index=true');
The preceding example creates a composite B-tree spatial index, not an R-tree spatial index. Using a composite B-tree spatial index for point data can:
-
Improve the performance of spatial index creation.
-
Improve DML performance when performing concurrent DML from many Oracle sessions.
However, while composite B-tree spatial query performance is very fast, optimal spatial query performance may be obtained by using an R-tree spatial index on that data (especially with SPATIAL_VECTOR_ACCELERATION set to the recommended value of TRUE
).
The cbtree_index=true
keyword can used only for spatial index creation (CREATE INDEX statement). It cannot be used with ALTER INDEX or ALTER INDEX REBUILD.
The SDO_JOIN operator is not supported when a composite B-tree spatial index is used
5.1.4 Creating a Cross-Schema Index
You can create a spatial index on a table that is not in your schema. Assume that user B wants to create a spatial index on column GEOMETRY in table T1 under user A's schema. Follow these steps:
5.1.5 Using Partitioned Spatial Indexes
You can create a partitioned spatial index on a partitioned table. This section describes usage considerations specific to Oracle Spatial and Graph. For a detailed explanation of partitioned tables and partitioned indexes, see Oracle Database VLDB and Partitioning Guide.
A partitioned spatial index can provide the following benefits:
-
Reduced response times for long-running queries, because partitioning reduces disk I/O operations
-
Reduced response times for concurrent queries, because I/O operations run concurrently on each partition
-
Easier index maintenance, because of partition-level create and rebuild operations
Indexes on partitions can be rebuilt without affecting the queries on other partitions, and storage parameters for each local index can be changed independent of other partitions.
-
Parallel query on multiple partition searching
The degree of parallelism is the value from the DEGREE column in the row for the index in the USER_INDEXES view (that is, the value specified or defaulted for the PARALLEL keyword with the CREATE INDEX, ALTER INDEX, or ALTER INDEX REBUILD statement).
-
Improved query processing in multiprocessor system environments
In a multiprocessor system environment, if a spatial operator is invoked on a table with partitioned spatial index and if multiple partitions are involved in the query, multiple processors can be used to evaluate the query. The number of processors used is determined by the degree of parallelism and the number of partitions used in evaluating the query.
The following restrictions apply to spatial index partitioning:
-
The partition key for spatial tables must be a scalar value, and must not be a spatial column.
-
Only range partitioning is supported on the underlying table. All other kinds of partitioning are not currently supported for partitioned spatial indexes.
To create a partitioned spatial index, you must specify the LOCAL keyword. (If you do not specify the LOCAL keyword, a nonpartitioned spatial index is created on the data in all table partitions.) The following example creates a partitioned spatial index:
CREATE INDEX counties_idx ON counties(geometry) INDEXTYPE IS MDSYS.SPATIAL_INDEX_V2 LOCAL;
In this example, the default values are used for the number and placement of index partitions, namely:
-
Index partitioning is based on the underlying table partitioning. For each table partition, a corresponding index partition is created.
-
Each index partition is placed in the default tablespace.
If you do specify parameters for individual partitions, the following considerations apply:
-
The storage characteristics for each partition can be the same or different for each partition. If they are different, it may enable parallel I/O (if the tablespaces are on different disks) and may improve performance.
-
The
sdo_indx_dims
value must be the same for all partitions. -
The
layer_gtype
parameter value (see Constraining Data to a Geometry Type) used for each partition may be different.
To override the default partitioning values, use a CREATE INDEX statement with the following general format:
CREATE INDEX <indexname> ON <table>(<column>) INDEXTYPE IS MDSYS.SPATIAL_INDEX_V2 [PARAMETERS ('<spatial-params>, <storage-params>')] LOCAL [( PARTITION <index_partition> PARAMETERS ('<spatial-params>, <storage-params>') [, PARTITION <index_partition> PARAMETERS ('<spatial-params>, <storage-params>')] )]
Queries can operate on partitioned tables to perform the query on only one partition. For example:
SELECT * FROM counties PARTITION(p1) WHERE ...<some-spatial-predicate>;
Querying on a selected partition may speed up the query and also improve overall throughput when multiple queries operate on different partitions concurrently.
When queries use a partitioned spatial index, the semantics (meaning or behavior) of spatial operators and functions is the same with partitioned and nonpartitioned indexes, except in the case of SDO_NN (nearest neighbor). With SDO_NN, the requested number of geometries is returned for each partition that is affected by the query. (See the description of the SDO_NN operator in Spatial Operators for more information.)
For example, if you request the 5 closest restaurants to a point and the spatial index has 4 partitions, SDO_NN returns up to 20 (5*4) geometries. In this case, you must use the ROWNUM pseudocolumn (here, WHERE ROWNUM <=5
) to return the 5 closest restaurants, and the ORDER BY clause to sort the results by distance in miles. Example 5-1 returns the 5 nearest restaurants from a partitioned spatial index.
See Also:
SDO_NN Examples for more examples of using the SDO_NN operator.
For a cross-schema query when a table has a partitioned spatial index, the user must be granted SELECT or READ privilege on both the spatial table and the index table (MDRT_xxx) for the spatial index that was created on the spatial table. For more information and an example, see "Cross-Schema Invocation of SDO_JOIN" in the Usage Notes for the SDO_JOIN operator.
Example 5-1 SDO_NN Query with Partitioned Spatial Index
SELECT * FROM ( SELECT r.name, r.location, SDO_NN_DISTANCE(1) distance_in_miles FROM restaurants_part_table r WHERE SDO_NN(r.location, MDSYS.SDO_GEOMETRY(2001,8307,MDSYS.SDO_POINT_TYPE(-110,35,Null),Null,Null), 'SDO_NUM_RES=5 distance=2 unit=MILE', 1) = 'TRUE' ORDER BY distance_in_miles ) WHERE ROWNUM<=5;
5.1.5.1 Creating a Local Partitioned Spatial Index
If you want to create a local partitioned spatial index, Oracle recommends that you use the procedure in this section instead of using the PARALLEL keyword, to avoid having to start over if the creation of any partition's index fails for any reason (for example, because the tablespace is full). Follow these steps:
If any of the ALTER INDEX statements fails, you do not need to rebuild any partitions for which the operation has successfully completed.
5.1.6 Exchanging Partitions Including Indexes
You can use the ALTER TABLE statement with the EXCHANGE PARTITION ... INCLUDING INDEXES clause to exchange a spatial table partition and its index partition with a corresponding table and its index. For information about exchanging partitions, see the description of the ALTER TABLE statement in Oracle Database SQL Language Reference.
This feature can help you to operate more efficiently in a number of situations, such as:
-
Bringing data into a partitioned table and avoiding the cost of index re-creation.
-
Managing and creating partitioned indexes. For example, the data could be divided into multiple tables. The index for each table could be built one after the other to minimize the memory and tablespace resources needed during index creation. Alternately, the indexes could be created in parallel in multiple sessions. The tables (along with the indexes) could then be exchanged with the partitions of the original data table.
-
Managing offline insert operations. New data can be stored in a temporary table and periodically exchanged with a new partition (for example, in a database with historical data).
To exchange partitions including indexes with spatial data and indexes, the two spatial indexes (one on the partition, the other on the table) must have the same dimensionality (sdo_indx_dims
value). If the indexes do not have the same dimensionality, an error is raised. The table data is exchanged, but the indexes are not exchanged and the indexes are marked as failed. To use the indexes, you must rebuild them
5.1.7 Export and Import Considerations with Spatial Indexes and Data
If you use the Export utility to export tables with spatial data, the behavior of the operation depends on whether or not the spatial data has been spatially indexed:
-
If the spatial data has not been spatially indexed, the table data is exported. However, you must update the USER_SDO_GEOM_METADATA view with the appropriate information on the target system.
-
If the spatial data has been spatially indexed, the table data is exported, the appropriate information is inserted into the USER_SDO_GEOM_METADATA view on the target system, and the spatial index is built on the target system. However, if the insertion into the USER_SDO_GEOM_METADATA view fails (for example, if there is already a USER_SDO_GEOM_METADATA entry for the spatial layer), the spatial index is not built.
If you use the Import utility to import data that has been spatially indexed, the following considerations apply:
-
If the index on the exported data was created with a
TABLESPACE
clause and if the specified tablespace does not exist in the database at import time, the index is not built. (This is different from the behavior with other Oracle indexes, where the index is created in the user's default tablespace if the tablespace specified for the original index does not exist at import time.) -
If the import operation must be done by a privileged database user, and if the
FROMUSER
andTOUSER
format is used, theTOUSER
user must be granted the CREATE TABLE and CREATE SEQUENCE privileges before the import operation, as shown in the following example (and enter the password for the SYSTEM account when prompted):sqlplus system SQL> grant CREATE TABLE, CREATE SEQUENCE to CHRIS; SQL> exit; imp system file=spatl_data.dmp fromuser=SCOTT touser=CHRIS
For information about using the Export and Import utilities, see Oracle Database Utilities.
5.1.8 Distributed and Oracle XA Transactions Not Supported with R-Tree Spatial Indexes
The use of R-tree spatial indexes is not supported in any distributed or Oracle XA transaction. The use of distributed transactions in conjunction with R-tree spatial indexes may result in the current transaction being terminated and rolled back with the following exception:
ORA-13078: "cannot commit a distributed transaction involving spatial index"
For example, do not use DML operations on any table containing R-tree spatial indexes within distributed or Oracle XA transactions, because R-tree spatial index use in this context is not officially supported.
Note:
A composite B-tree index on point data (described in Creating a Composite B-tree Spatial Index on Points) is supported for distributed and Oracle XA transactions.
For more information about distributed transactions, see Oracle Database Administrator's Guide.
5.1.9 Enabling Access to Spatial Index Statistics
The Oracle Database optimizer collects statistics that describe details about the database and its objects. Statistics are critical to the optimizer's ability to pick the best execution plan for a SQL statement. For more information about optimizer statistics, see Oracle Database SQL Tuning Guide.
To be able to use procedures such as DBMS_STATS.GATHER_INDEX_STATS
and DBMS_STATS.GATHER_SCHEMA_STATS
to gather index statistics related to spatial indexes, the CREATE TABLE privilege must be granted to all database users that will perform the statistics collection.
When you run ANALYZE INDEX on a spatial domain index for a different schema (user), the user performing the ANALYZE operation needs the following privileges:
-
CREATE ANY TABLE to create missing temporary tables
-
DROP ANY TABLE to truncate or remove existing temporary tables
If the statistics are successfully collected, a table with the name in the form MDXT_%objID% will be created for each index for which the statistics are collected. Each such table will be populated with spatial statistics data, 512 rows by default.
5.1.10 Rollback Segments and Sort Area Size
This section applies only if you (or the database administrator) are not using the automatic undo management feature or the PGA memory management feature, or both, of Oracle Database. Automatic memory management and PGA memory management are enabled by default, and their use is highly recommended. For explanations of these features, see:
-
The section about automatic undo management and undo segments in Oracle Database Concepts
-
The section about PGA memory management in Oracle Database Concepts
If you are not using automatic undo management and if the rollback segment is not large enough, an attempt to create a spatial index will fail. The rollback segment should be 100*n bytes, where n is the number of rows of data to be indexed. For example, if the table contains 1 million (1,000,000) rows, the rollback segment size should be 100,000,000 (100 million) bytes.
To ensure an adequate rollback segment, or if you have tried to create a spatial index and received an error that a rollback segment cannot be extended, review (or have a DBA review) the size and structure of the rollback segments. Create a public rollback segment of the appropriate size, and place that rollback segment online. In addition, ensure that any small inappropriate rollback segments are placed offline during large spatial index operations.
If you are not using the PGA memory management feature, the database parameter SORT_AREA_SIZE affects the amount of time required to create the index. The SORT_AREA_SIZE value is the maximum amount, in bytes, of memory to use for a sort operation. The optimal value depends on the database size, but a good guideline is to make it at least 1 million bytes when you create a spatial index. To change the SORT_AREA_SIZE value, use the ALTER SESSION statement. For example, to change the value to 20 million bytes:
ALTER SESSION SET SORT_AREA_SIZE = 20000000;
5.2 Querying Spatial Data
The structures of a spatial layer are used to resolve spatial queries and spatial joins.
Spatial and Graph uses a two-tier query model with primary and secondary filter operations to resolve spatial queries and spatial joins, as explained in Query Model. The term two-tier indicates that two distinct operations are performed to resolve queries. If both operations are performed, the exact result set is returned.
You cannot append a database link (dblink) name to the name of a spatial table in a query if a spatial index is defined on that table.
This section also contains the following topics.
5.2.1 Spatial Query
In a spatial R-tree index, each geometry is represented by its minimum bounding rectangle (MBR), as explained in R-Tree Indexing. Consider the following layer containing several objects in Figure 5-1. Each object is labeled with its geometry name (geom_1 for the line string, geom_2 for the four-sided polygon, geom_3 for the triangular polygon, and geom_4 for the ellipse), and the MBR around each object is represented by a dashed line.
A typical spatial query is to request all objects that lie within a query window, that is, a defined fence or window. A dynamic query window refers to a rectangular area that is not defined in the database, but that must be defined before it is used. Figure 5-2 shows the same geometries as in Figure 5-1, but adds a query window represented by the heavy dotted-line box.
In Figure 5-2, the query window covers parts of geometries geom_1 and geom_2, as well as part of the MBR for geom_3 but none of the actual geom_3 geometry. The query window does not cover any part of the geom_4 geometry or its MBR.
Topics:
5.2.1.1 Primary Filter Operator
The SDO_FILTER operator, described in Spatial Operators , implements the primary filter portion of the two-step process involved in the Oracle Spatial and Graph query processing model. The primary filter uses the index data to determine only if a set of candidate object pairs may interact. Specifically, the primary filter checks to see if the MBRs of the candidate objects interact, not whether the objects themselves interact. The SDO_FILTER operator syntax is as follows:
SDO_FILTER(geometry1 SDO_GEOMETRY, geometry2 SDO_GEOMETRY, param VARCHAR2)
In the preceding syntax:
-
geometry1
is a column of type SDO_GEOMETRY in a table. This column must be spatially indexed. -
geometry2
is an object of type SDO_GEOMETRY. This object may or may not come from a table. If it comes from a table, it may or may not be spatially indexed. -
param
is an optional string of type VARCHAR2. It can specify either or both of themin_resolution
andmax_resolution
keywords.
The following examples perform a primary filter operation only (with no secondary filter operation). They will return all the geometries shown in Figure 5-2 that have an MBR that interacts with the query window. The result of the following examples are geometries geom_1, geom_2, and geom_3.
Example 5-2 performs a primary filter operation without inserting the query window into a table. The window will be indexed in memory and performance will be very good.
Example 5-2 Primary Filter with a Temporary Query Window
SELECT A.Feature_ID FROM TARGET A WHERE sdo_filter(A.shape, SDO_geometry(2003,NULL,NULL, SDO_elem_info_array(1,1003,3), SDO_ordinate_array(x1,y1, x2,y2)) ) = 'TRUE';
In Example 5-2, (x1,y1)
and (x2,y2)
are the lower-left and upper-right corners of the query window.
In Example 5-3, a transient instance of type SDO_GEOMETRY was constructed for the query window instead of specifying the window parameters in the query itself.
Example 5-3 Primary Filter with a Transient Instance of the Query Window
SELECT A.Feature_ID FROM TARGET A WHERE sdo_filter(A.shape, :theWindow) = 'TRUE';
Example 5-4 assumes the query window was inserted into a table called WINDOWS, with an ID of WINS_1.
Example 5-4 Primary Filter with a Stored Query Window
SELECT A.Feature_ID FROM TARGET A, WINDOWS B WHERE B.ID = 'WINS_1' AND sdo_filter(A.shape, B.shape) = 'TRUE';
If the B.SHAPE column is not spatially indexed, the SDO_FILTER operator indexes the query window in memory and performance is very good.
5.2.1.2 Primary and Secondary Filter Operator
The SDO_RELATE operator, described in Spatial Operators , performs both the primary and secondary filter stages when processing a query. The secondary filter ensures that only candidate objects that actually interact are selected. This operator can be used only if a spatial index has been created on two dimensions of data. The syntax of the SDO_RELATE operator is as follows:
SDO_RELATE(geometry1 SDO_GEOMETRY, geometry2 SDO_GEOMETRY, param VARCHAR2)
In the preceding syntax:
-
geometry1
is a column of type SDO_GEOMETRY in a table. This column must be spatially indexed. -
geometry2
is an object of type SDO_GEOMETRY. This object may or may not come from a table. If it comes from a table, it may or may not be spatially indexed. -
param
is a quoted string with themask
keyword and a valid mask value, and optionally either or both of themin_resolution
andmax_resolution
keywords, as explained in the documentation for the SDO_RELATE operator in Spatial Operators .
The following examples perform both primary and secondary filter operations. They return all the geometries in Figure 5-2 that lie within or overlap the query window. The result of these examples is objects geom_1 and geom_2.
Example 5-5 performs both primary and secondary filter operations without inserting the query window into a table. The window will be indexed in memory and performance will be very good.
Example 5-5 Secondary Filter Using a Temporary Query Window
SELECT A.Feature_ID FROM TARGET A WHERE sdo_relate(A.shape, SDO_geometry(2003,NULL,NULL, SDO_elem_info_array(1,1003,3), SDO_ordinate_array(x1,y1, x2,y2)), 'mask=anyinteract') = 'TRUE';
In Example 5-5, (x1,y1)
and (x2,y2)
are the lower-left and upper-right corners of the query window.
Example 5-6 assumes the query window was inserted into a table called WINDOWS, with an ID value of WINS_1.
Example 5-6 Secondary Filter Using a Stored Query Window
SELECT A.Feature_ID FROM TARGET A, WINDOWS B WHERE B.ID = 'WINS_1' AND sdo_relate(A.shape, B.shape, 'mask=anyinteract') = 'TRUE';
If the B.SHAPE column is not spatially indexed, the SDO_RELATE operator indexes the query window in memory and performance is very good.
5.2.1.3 Within-Distance Operator
The SDO_WITHIN_DISTANCE operator, described in Spatial Operators , is used to determine the set of objects in a table that are within n distance units from a reference object. This operator can be used only if a spatial index has been created on two dimensions of data. The reference object may be a transient or persistent instance of SDO_GEOMETRY, such as a temporary query window or a permanent geometry stored in the database. The syntax of the operator is as follows:
SDO_WITHIN_DISTANCE(geometry1 SDO_GEOMETRY, aGeom SDO_GEOMETRY, params VARCHAR2);
In the preceding syntax:
-
geometry1
is a column of type SDO_GEOMETRY in a table. This column must be spatially indexed. -
aGeom
is an instance of type SDO_GEOMETRY. -
params
is a quoted string of keyword value pairs that determines the behavior of the operator. See the SDO_WITHIN_DISTANCE operator in Spatial Operators for a list of parameters.
The following example selects any objects within 1.35 distance units from the query window:
SELECT A.Feature_ID FROM TARGET A WHERE SDO_WITHIN_DISTANCE( A.shape, :theWindow, 'distance=1.35') = 'TRUE';
The distance units are based on the geometry coordinate system in use. If you are using a geodetic coordinate system, the units are meters. If no coordinate system is used, the units are the same as for the stored data.
The SDO_WITHIN_DISTANCE operator is not suitable for performing spatial joins. That is, a query such as Find all parks that are within 10 distance units from coastlines will not be processed as an index-based spatial join of the COASTLINES and PARKS tables. Instead, it will be processed as a nested loop query in which each COASTLINES instance is in turn a reference object that is buffered, indexed, and evaluated against the PARKS table. Thus, the SDO_WITHIN_DISTANCE operation is performed n times if there are n rows in the COASTLINES table.
For non-geodetic data, there is an efficient way to accomplish a spatial join that involves buffering all geometries of a layer. This method does not use the SDO_WITHIN_DISTANCE operator. First, create a new table COSINE_BUFS as follows:
CREATE TABLE cosine_bufs UNRECOVERABLE AS SELECT SDO_BUFFER (A.SHAPE, B.DIMINFO, 1.35) FROM COSINE A, USER_SDO_GEOM_METADATA B WHERE TABLE_NAME='COSINES' AND COLUMN_NAME='SHAPE';
Next, create a spatial index on the SHAPE column of COSINE_BUFS. Then you can perform the following query:
SELECT /*+ ordered */ a.gid, b.gid FROM TABLE(SDO_JOIN('PARKS', 'SHAPE', 'COSINE_BUFS', 'SHAPE', 'mask=ANYINTERACT')) c, parks a, cosine_bufs b WHERE c.rowid1 = a.rowid AND c.rowid2 = b.rowid;
5.2.1.4 Nearest Neighbor Operator
The SDO_NN operator, described in Spatial Operators , is used to identify the nearest neighbors for a geometry. This operator can be used only if a spatial index has been created on two dimensions of data. The syntax of the operator is as follows:
SDO_NN(geometry1 SDO_GEOMETRY, geometry2 SDO_GEOMETRY, param VARCHAR2 [, number NUMBER]);
In the preceding syntax:
-
geometry1
is a column of type SDO_GEOMETRY in a table. This column must be spatially indexed. -
geometry2
is an instance of type SDO_GEOMETRY. -
param
is a quoted string of keyword-value pairs that can determine the behavior of the operator, such as how many nearest neighbor geometries are returned. See the SDO_NN operator in Spatial Operators for information about this parameter. -
number
is the same number used in the call to SDO_NN_DISTANCE. Use this only if the SDO_NN_DISTANCE ancillary operator is included in the call to SDO_NN. See the SDO_NN operator in Spatial Operators for information about this parameter.
The following example finds the two objects from the SHAPE column in the COLA_MARKETS table that are closest to a specified point (10,7). (Note the use of the optimizer hint in the SELECT statement, as explained in the Usage Notes for the SDO_NN operator in Spatial Operators .)
SELECT /*+ INDEX(cola_markets cola_spatial_idx) */ c.mkt_id, c.name FROM cola_markets c WHERE SDO_NN(c.shape, SDO_geometry(2001, NULL, SDO_point_type(10,7,NULL), NULL, NULL), 'sdo_num_res=2') = 'TRUE';
5.2.1.5 Spatial Functions
Spatial and Graph also supplies functions for determining relationships between geometries, finding information about single geometries, changing geometries, and combining geometries. These functions all take into account two dimensions of source data. If the output value of these functions is a geometry, the resulting geometry will have the same dimensionality as the input geometry, but only the first two dimensions will accurately reflect the result of the operation.
5.2.2 Spatial Join
A spatial join is the same as a regular join except that the predicate involves a spatial operator. In Spatial and Graph, a spatial join takes place when you compare all geometries of one layer to all geometries of another layer. This is unlike a query window, which compares a single geometry to all geometries of a layer.
Spatial joins can be used to answer questions such as Which highways cross national parks?
The following table structures illustrate how the join would be accomplished for this example:
PARKS( GID VARCHAR2(32), SHAPE SDO_GEOMETRY) HIGHWAYS( GID VARCHAR2(32), SHAPE SDO_GEOMETRY)
To perform a spatial join, use the SDO_JOIN operator, which is described in Spatial Operators . The following spatial join query, to list the GID column values of highways and parks where a highway interacts with a park, performs a primary filter operation only ('mask=FILTER'
), and thus it returns only approximate results:
SELECT /*+ ordered */ a.gid, b.gid FROM TABLE(SDO_JOIN('PARKS', 'SHAPE', 'HIGHWAYS', 'SHAPE', 'mask=FILTER')) c, parks a, highways b WHERE c.rowid1 = a.rowid AND c.rowid2 = b.rowid;
Note:
The SDO_JOIN operator is not supported when a composite B-tree spatial index is used.
The following spatial join query requests the same information as in the preceding example, but it performs both primary and secondary filter operations ('mask=ANYINTERACT'
), and thus it returns exact results:
SELECT /*+ ordered */ a.gid, b.gid FROM TABLE(SDO_JOIN('PARKS', 'SHAPE', 'HIGHWAYS', 'SHAPE', 'mask=ANYINTERACT')) c, parks a, highways b WHERE c.rowid1 = a.rowid AND c.rowid2 = b.rowid;
5.2.3 Data and Index Dimensionality, and Spatial Queries
The elements of a spatial query can, in theory, have the following dimensionality:
-
The base table geometries (or
geometry1
in spatial operator formats) can have two, three, or more dimensions. -
The spatial index created on the base table (or
geometry1
) can be two-dimensional or three-dimensional. -
The query window (or
geometry2
in spatial operator formats) can have two, three, or more dimensions.
Some combinations of dimensionality among the three elements are supported and some are not. Table 5-1 explains what happens with the possible combinations involving two and three dimensions.
Table 5-1 Data and Index Dimensionality, and Query Support
Base Table (geometry1) Dimensionality | Spatial Index Dimensionality | Query Window (geometry2) Dimensionality | Query Result |
---|---|---|---|
2-dimensional |
2-dimensional |
2-dimensional |
Performs a two-dimensional query. |
2-dimensional |
2-dimensional |
3-dimensional |
Supported if the query window has an appropriate SDO_GTYPE value less than 3008. |
2-dimensional |
3-dimensional |
2-dimensional |
Not supported: 3D index not permitted on 2D data. |
2-dimensional |
3-dimensional |
3-dimensional |
Not supported: 3D index not permitted on 2D data. |
3-dimensional |
2-dimensional |
2-dimensional |
Ignores the third (Z) dimension in each base geometry and performs a two-dimensional query. |
3-dimensional |
2-dimensional |
3-dimensional |
Supported if the query window has an appropriate SDO_GTYPE value less than 3008. |
3-dimensional |
3-dimensional |
2-dimensional |
Converts the 2D query window to a 3D window with zero Z values and performs a three-dimensional query. |
3-dimensional |
3-dimensional |
3-dimensional |
Performs a three-dimensional query. |