Chapter 14

Managing Column Group Statistics

SELECT DBMS_STATS.REPORT_COL_USAGE(user, 'customers_test')

FROM DUAL;

The report appears below:

LEGEND:
.......

EQ : Used in single table EQuality predicate

RANGE : Used in single table RANGE predicate
LIKE : Used in single table LIKE predicate
NULL : Used in single table is (not) NULL predicate
EQ_JOIN : Used in EQuality JOIN predicate
NONEQ_JOIN : Used in NON EQuality JOIN predicate
FILTER : Used in single table FILTER predicate
JOIN : Used in JOIN predicate
GROUP_BY : Used in GROUP BY expression
....................................................................
....

####################################################################
####

COLUMN USAGE REPORT FOR SH.CUSTOMERS_TEST

.........................................

1. COUNTRY_ID : EQ
2. CUST_CITY : EQ
3. CUST_STATE_PROVINCE : EQ
4. (CUST_CITY, CUST_STATE_PROVINCE,
COUNTRY_ID) : FILTER
5. (CUST_STATE_PROVINCE, COUNTRY_ID) : GROUP_BY
####################################################################
####

In the preceding report, the first three columns were used in equality predicates in
the first monitored query:

...
WHERE cust_city = 'Los Angeles'
AND cust_state_province = 'CA'
AND country_id = 52790;

All three columns appeared in the same WHERE clause, so the report shows them
as a group filter. In the second query, two columns appeared in the GROUP BY
clause, so the report labels them as GROUP_BY. The sets of columns in the FILTER
and GROUP_BY report are candidates for column groups.

14-8
 Chapter 14
Managing Column Group Statistics

See Also:

• "Managing SQL Tuning Sets"

• Oracle Database PL/SQL Packages and Types Reference to learn about the
DBMS_STATS package

14.1.3 Creating Column Groups Detected During Workload Monitoring

You can use the DBMS_STATS.CREATE_EXTENDED_STATS function to create column groups that
were detected previously by executing DBMS_STATS.SEED_COL_USAGE.

Assumptions
This tutorial assumes that you have performed the steps in "Detecting Useful Column Groups
for a Specific Workload".

To create column groups:

1. Create column groups for the customers_test table based on the usage information
captured during the monitoring window.
For example, run the following query:

SELECT DBMS_STATS.CREATE_EXTENDED_STATS(user, 'customers_test') FROM DUAL;

Sample output appears below:

########################################################################
EXTENSIONS FOR SH.CUSTOMERS_TEST
................................
1. (CUST_CITY, CUST_STATE_PROVINCE,
COUNTRY_ID) :SYS_STUMZ$C3AIHLPBROI#SKA58H_N
created
2. (CUST_STATE_PROVINCE, COUNTRY_ID):SYS_STU#S#WF25Z#QAHIHE#MOFFMM_
created
########################################################################

The database created two column groups for customers_test: one column group for the
filter predicate and one group for the GROUP BY operation.
2. Regather table statistics.
Run GATHER_TABLE_STATS to regather the statistics for customers_test:

EXEC DBMS_STATS.GATHER_TABLE_STATS(user,'customers_test');

3. As user sh, run explain plans for two queries in the workload.

14-9
 Chapter 14
Managing Column Group Statistics

Check the USER_TAB_COL_STATISTICS view to determine which additional statistics

were created by the database:

SELECT COLUMN_NAME, NUM_DISTINCT, HISTOGRAM

FROM USER_TAB_COL_STATISTICS
WHERE TABLE_NAME = 'CUSTOMERS_TEST'
ORDER BY 1;

Partial sample output appears below:

CUST_CITY 620 HEIGHT BALANCED

...
SYS_STU#S#WF25Z#QAHIHE#MOFFMM_ 145 NONE
SYS_STUMZ$C3AIHLPBROI#SKA58H_N 620 HEIGHT BALANCED

This example shows the two column group names returned from the
DBMS_STATS.CREATE_EXTENDED_STATS function. The column group created on
CUST_CITY, CUST_STATE_PROVINCE, and COUNTRY_ID has a height-balanced
histogram.
4. Explain the plans again.
The following examples show the explain plans for two queries on the
customers_test table:

EXPLAIN PLAN FOR

SELECT *
FROM customers_test
WHERE cust_city = 'Los Angeles'
AND cust_state_province = 'CA'
AND country_id = 52790;

SELECT PLAN_TABLE_OUTPUT
FROM TABLE(DBMS_XPLAN.DISPLAY('plan_table', null,'basic rows'));

EXPLAIN PLAN FOR

SELECT country_id, cust_state_province, count(cust_city)
FROM customers_test
GROUP BY country_id, cust_state_province;

SELECT PLAN_TABLE_OUTPUT
FROM TABLE(DBMS_XPLAN.DISPLAY('plan_table', null,'basic rows'));

The new plans show more accurate cardinality estimates:

----------------------------------------------------
| Id | Operation | Name | Rows |
----------------------------------------------------
| 0 | SELECT STATEMENT | | 1093 |
| 1 | TABLE ACCESS FULL| CUSTOMERS_TEST | 1093 |
----------------------------------------------------

8 rows selected.

14-10
 Chapter 14
Managing Column Group Statistics

Plan hash value: 3050654408

-----------------------------------------------------
| Id | Operation | Name | Rows |
-----------------------------------------------------
| 0 | SELECT STATEMENT | | 145 |
| 1 | HASH GROUP BY | | 145 |
| 2 | TABLE ACCESS FULL| CUSTOMERS_TEST | 55500 |
-----------------------------------------------------
9 rows selected.

See Also:
Oracle Database PL/SQL Packages and Types Reference to learn about the
DBMS_STATS package

14.1.4 Creating and Gathering Statistics on Column Groups Manually

In some cases, you may know the column group that you want to create.
The METHOD_OPT argument of the DBMS_STATS.GATHER_TABLE_STATS function can create and
gather statistics on a column group automatically. You can create a new column group by
specifying the group of columns using FOR COLUMNS.

Assumptions
This tutorial assumes the following:
• You want to create a column group for the cust_state_province and country_id
columns in the customers table in sh schema.
• You want to gather statistics (including histograms) on the entire table and the new
column group.

To create a column group and gather statistics for this group:

1. In SQL*Plus, log in to the database as the sh user.
2. Create the column group and gather statistics.
For example, execute the following PL/SQL program:

BEGIN
DBMS_STATS.GATHER_TABLE_STATS( 'sh','customers',
METHOD_OPT => 'FOR ALL COLUMNS SIZE SKEWONLY ' ||
'FOR COLUMNS SIZE SKEWONLY (cust_state_province,country_id)' );
END;
/

14-11
 Chapter 14
Managing Column Group Statistics

See Also:
Oracle Database PL/SQL Packages and Types Reference to learn about the
DBMS_STATS.GATHER_TABLE_STATS procedure

14.1.5 Displaying Column Group Information

To obtain the name of a column group, use the
DBMS_STATS.SHOW_EXTENDED_STATS_NAME function or a database view.

You can also use views to obtain information such as the number of distinct values,
and whether the column group has a histogram.

Assumptions
This tutorial assumes the following:
• You created a column group for the cust_state_province and country_id
columns in the customers table in sh schema.
• You want to determine the column group name, the number of distinct values, and
whether a histogram has been created for a column group.

To monitor a column group:

1. Start SQL*Plus and connect to the database as the sh user.
2. To determine the column group name, do one of the following.
• Execute the SHOW_EXTENDED_STATS_NAME function.
For example, run the following PL/SQL program:

SELECT SYS.DBMS_STATS.SHOW_EXTENDED_STATS_NAME( 'sh','customers',

'(cust_state_province,country_id)' ) col_group_name
FROM DUAL;

The output is similar to the following:

COL_GROUP_NAME
----------------
SYS_STU#S#WF25Z#QAHIHE#MOFFMM_

• Query the USER_STAT_EXTENSIONS view.

For example, run the following query:

SELECT EXTENSION_NAME, EXTENSION

FROM USER_STAT_EXTENSIONS
WHERE TABLE_NAME='CUSTOMERS';

EXTENSION_NAME EXTENSION
-----------------------------------------------------------------
---

14-12
 Chapter 14
Managing Column Group Statistics

SYS_STU#S#WF25Z#QAHIHE#MOFFMM_ ("CUST_STATE_PROVINCE","COUNTRY_ID")

3. Query the number of distinct values and find whether a histogram has been created for a
column group.
For example, run the following query:

SELECT e.EXTENSION col_group, t.NUM_DISTINCT, t.HISTOGRAM

FROM USER_STAT_EXTENSIONS e, USER_TAB_COL_STATISTICS t
WHERE e.EXTENSION_NAME=t.COLUMN_NAME
AND e.TABLE_NAME=t.TABLE_NAME
AND t.TABLE_NAME='CUSTOMERS';

COL_GROUP NUM_DISTINCT HISTOGRAM

-------------------------------------------------------------------
("COUNTRY_ID","CUST_STATE_PROVINCE") 145 FREQUENCY

See Also:
Oracle Database PL/SQL Packages and Types Reference to learn about the
DBMS_STATS.SHOW_EXTENDED_STATS_NAME function

14.1.6 Dropping a Column Group

Use the DBMS_STATS.DROP_EXTENDED_STATS function to delete a column group from a table.

Assumptions
This tutorial assumes the following:
• You created a column group for the cust_state_province and country_id columns in
the customers table in sh schema.
• You want to drop the column group.

To drop a column group:

1. Start SQL*Plus and connect to the database as the sh user.
2. Drop the column group.
For example, the following PL/SQL program deletes a column group from the customers
table:

BEGIN
DBMS_STATS.DROP_EXTENDED_STATS( 'sh', 'customers',
'(cust_state_province, country_id)' );
END;
/

14-13