Advantage of ON PREBUILT in Materialized views

Can someone please provide me the advantages of using the ON PREBUILT option over regular method in REFRESH COMPLETE materialized views
I am new to this Mviews and my requirement is to reduce the load on DB in the environment where both the production and MViews refresh is happening in the same DB.
If you have any other idea of refreshing the Mviews in such a scenario please provide your valuabale comments
Edited by: 985844 on Feb 3, 2013 9:30 PM

Welcome to the forum!
Whenever you post provide your 4 digit Oracle version.
>
Can someone please provide me the advantages of using the ON PREBUILT option over regular method in REFRESH COMPLETE materialized views
I am new to this Mviews and my requirement is to reduce the load on DB in the environment where both the production and MViews refresh is happening in the same DB.
If you have any other idea of refreshing the Mviews in such a scenario please provide your valuabale comments
>
You are confusing the creation of an MV with the refresh of the MV. That clause is used when you first create the MV. If you use an existing table as the source it is faster. If you don't use the ON PREBUILT clause then Oracle has to create a table to use and then populate it with data.
You would not recreate the MV itself to do a refresh. When you convert an existing table to an MV your users will not be manually altering the data in the MV you will refresh it either manually or automatically.
See that clause in the Oracle docs
http://docs.oracle.com/cd/B28359_01/server.111/b28286/statements_6002.htm
>
ON PREBUILT TABLE Clause
The ON PREBUILT TABLE clause lets you register an existing table as a preinitialized materialized view. This clause is particularly useful for registering large materialized views in a data warehousing environment. The table must have the same name and be in the same schema as the resulting materialized view.
If the materialized view is dropped, then the preexisting table reverts to its identity as a table.
>
You also said
>
I am new to this Mviews and my requirement is to reduce the load on DB in the environment where both the production and MViews refresh is happening in the same DB.
>
How did you determine that the MV refresh is what is causing any problem that you have.
Remember, if it ain't broke don't fix it. The first step is to determine exactly what the problem is.

Similar Messages

Advantage of Materialized view on prebuilt table

Could someone tell me the advantage of Materialized view on prebuilt table?. I'm unable to understand the concept from the Oracle documentation. I need to know the answers for following questions.
1) Is the data stored in table and MV same?. Does the query retrive data from table or MV?. The query internally uses either table or a view as the names of both are same.
2) Our DSS application is generating complicated queries and executing for long time. Is there any way I can optimize those queries using MV without rewriting the code in the application.

It's a roughly analgous problem to figuring out what set of indexes to create to improve the performance of an application-- you need to understand the various queries, how the application is accessing data, and you need to balance a variety of competing needs in order to come up with a reasonable set of indexes.
Fundamentally, Oracle can only use a materialized view to satisfy a query if a human could use just the data in the materialized view to answer the question. That generally means that the materialized view is going to have to be aggregated at a lower level or at the same level as the query. A materialized view that aggregates sales by day can be used for queries by year but a materialized view that aggregates sales by year cannot be used for queries that get sales by day. A materiailized view that aggregates sales by vendor and product can be used for queries that aggregate by vendor or by product, but a materialized view that aggregates by product cannot be used in a query that aggregates by product for a particular vendor.
You'll have to balance what materialized views are ideal for a particular query, what materialized views are sufficient for a particular set of queries, and how to balance the space, managability, and refresh performance requirements between creating lots of somewhat redundant materialized views that provide optimal query performance at the cost of a lot of disk and a large refresh window vs creating fewer, more general materialized views that consume less disk and can be refreshed faster but that provide less query performance boosts.
Justin

Materialized views on prebuilt tables - query rewrite

Hi Everyone,
I am currently counting on implementing the query rewrite functionality via materialized views to leverage existing aggregated tables.
Goal*: to use aggregate-awareness for our queries
How*: by creating views on existing aggregates loaded via ETL (+CREATE MATERIALIZED VIEW xxx on ON PREBUILT TABLE ENABLE QUERY REWRITE+)
Advantage*: leverage oracle functionalities + render logical model simpler (no aggregates)
Disadvantage*: existing ETL's need to be written as SQL in view creation statement --> aggregation rule exists twice (once on db, once in ETL)
Issue*: Certain ETL's are quite complex via lookups, functions, ... --> might create overy complex SQLs in view creation statements
My question: is there a way around the issue described? (I'm assuming the SQL in the view creation is necessary for oracle to know when an aggregate can be used)
Best practices & shared experiences are welcome as well of course
Kind regards,
Peter

streefpo wrote:
I'm still in the process of testing, but the drops should not be necessary.
Remember: The materialized view is nothing but a definition - the table itself continues to exist as before.
So as long as the definition doesn't change (added column, changed calculation, ...), the materialized view doesn't need to be re-created. (as the data is not maintained by Oracle)Thanks for reminding me but if you find a documented approach I will be waiting because this was the basis of my argument from the beginning.
SQL> select * from v$version ;
BANNER
Oracle Database 11g Enterprise Edition Release 11.2.0.1.0 - Production
PL/SQL Release 11.2.0.1.0 - Production
CORE     11.2.0.1.0     Production
TNS for Linux: Version 11.2.0.1.0 - Production
NLSRTL Version 11.2.0.1.0 - Production
SQL> desc employees
Name                                                                                            Null?    Type
EMPLOYEE_ID                                                                                     NOT NULL NUMBER(6)
FIRST_NAME                                                                                               VARCHAR2(20)
LAST_NAME                                                                                       NOT NULL VARCHAR2(25)
EMAIL                                                                                           NOT NULL VARCHAR2(25)
PHONE_NUMBER                                                                                             VARCHAR2(20)
HIRE_DATE                                                                                       NOT NULL DATE
JOB_ID                                                                                          NOT NULL VARCHAR2(10)
SALARY                                                                                                   NUMBER(8,2)
COMMISSION_PCT                                                                                           NUMBER(2,2)
MANAGER_ID                                                                                               NUMBER(6)
DEPARTMENT_ID                                                                                            NUMBER(4)
SQL> select count(*) from employees ;
COUNT(*)
       107
SQL> create table mv_table nologging as select department_id, sum(salary) as totalsal from employees group by department_id ;
Table created.
SQL> desc mv_table
Name                                                                                            Null?    Type
DEPARTMENT_ID                                                                                            NUMBER(4)
TOTALSAL                                                                                                 NUMBER
SQL> select count(*) from mv_table ;
COUNT(*)
        12
SQL> create materialized view mv_table on prebuilt table with reduced precision enable query rewrite as select department_id, sum(salary) as totalsal from employees group by department_id ;
Materialized view created.
SQL> select count(*) from mv_table ;
COUNT(*)
        12
SQL> select object_name, object_type from user_objects where object_name = 'MV_TABLE' ;
OBJECT_NAME                                                                                                                      OBJECT_TYPE
MV_TABLE                                                                                                                         TABLE
MV_TABLE                                                                                                                         MATERIALIZED VIEW
SQL> insert into mv_table values (999, 100) ;
insert into mv_table values (999, 100)
ERROR at line 1:
ORA-01732: data manipulation operation not legal on this view
SQL> update mv_table set totalsal = totalsal * 1.1 where department_id = 10 ;
update mv_table set totalsal = totalsal * 1.1 where department_id = 10
ERROR at line 1:
ORA-01732: data manipulation operation not legal on this view
SQL> delete from mv_table where totalsal <= 10000 ;
delete from mv_table where totalsal <= 10000
ERROR at line 1:
ORA-01732: data manipulation operation not legal on this view While investigating for this thread I actually made my own question redundant as the answer became gradually clear:
When using complex ETL's, I just need to make sure the complexity is located in the ETL loading the detailed table, not the aggregate
I'll try to clarify through an example:
- A detailed Table DET_SALES exists with Sales per Day, Store & Product
- An aggregated table AGG_SALES_MM exists with Sales, SalesStore per Month, Store & Product
- An ETL exists to load AGG_SALES_MM where Sales = SUM(Sales) & SalesStore = (SUM(Sales) Across Store)
--> i.e. the SalesStore measure will be derived out of a lookup
- A (Prebuilt) Materialized View will exist with the same column definitions as the ETL
--> to allow query-rewrite to know when to access the table
My concern was how to include the SalesStore in the materialized view definition (--> complex SQL!)
--> I should actually include SalesStore in the DET_SALES table, thus:
- including the 'Across Store' function in the detailed ETL
- rendering my Aggregation ETL into a simple GROUP BY
- rendering my materialized view definition into a simple GROUP BY as wellNot sure how close your example is to your actual problem. Also don't know if you are doing an incremental/complete data load and the data volume.
But the "SalesStore = (SUM(Sales) Across Store)" can be derived from the aggregated MV using analytical function. One can just create a normal view on top of MV for querying. It is hard to believe that aggregating in detail table during ETL load is the best approach but what do I know?

Materialized view from prebuilt table doesn't work with spatial types?

Hello, I'm trying to build a materialized view of a table using the prebuilt option and a pre-built table.
Oracle gives me an ORA-32304 error, saying it can't do this with user-defined types. The original table has no user-defined types, but does use an sdo_geometry column. Is this what it's complaining about?
Now I can sort of understand this, but here's my real problem: the materialized view I'm creating is comprised of a subset of the columns from the original table, but not the geometry column. Is Oracle right in refusing my prebuilt request? Does anyone know of a way around this (besides creating the MV from scratch without prebuilt).
I've successfully created an MV on another table, which doesn't have a spatial column, using the prebuilt option and a subset of columns.
(I'm using Oracle 11.2.0.2.0 on both master and slave databases)

Good news, everyone! =)
SAPwebIDE team fixed this issue with MMD template in SAPwebIDE v1.10.2. available on http://hanatrial.ondemand.com.
This "Bug" or "Feature" was presented in 1.8.x and 1.9.x SAPwebIDE (i've used local installation) and now it's gone in v1.10.2. Thank you, SAPwebIDE Team! =)
The difference between versions of MMD template is only in one file (fixed one is on the right):
Master2.controller.js
And here it is:
Now, only one question remains: HOWTO:SAPUI5 Fiori-like report. (mix control's value as key into binding context)
Best regards, ilia.

Query on Materialized View created on Prebuilt Table

Hi,
I am trying to explain the scenario below on Materialized View which I want to setup for our database.
I have created a materialized view on prebuilt table on target side on primary key. Now for purging history data from target side which are more than 5 years old I have executed the following steps:
1. Dropped the materialized view which was created previously on prebuilt table in target side.
2. Deleted the data from that target table which are more than 5 years old.
3. Now created the materialized view again on the prebuilt table.
Now, the problem which I am facing is that - if any changes happens in the source side during the above 3 steps on the target side those changed records were not captured even after successfully building materialized view again in the target side (i.e. after completing step-3 of above the intermediate changed records were not captured).
Can you please let me know exactly what I am doing wrong here and how can I achieve my intended result.
Regards,
Koushik

See matelink for doc id *252246.1*. The document id says: A materialized view was defined on a Table A. This table had a referential constraint defined against another table, Table B. This constraint was defined as 'ON DELETE CASCADE'. An 'ON DELETE CASCADE' constraint is not allowed on views but as the constraint was created on the table underlying the materialized view, Table A, it could be created although it would behave as a constraint on the view. The constraint existed for performance reasons, which is permitted, and was disabled when created but a general script had been run to enable all constraints. When a delete was performed on Table B the error above was reported although there was no view created on Table B.

Materialized View Stalesness Unusable

Hi,
I have a situation where a materialized view runs forever in SQL developer and doesn't comeback with a success message. I checked the properties of this Mview and it needed compiling (which i did). Post which the re run performance still doesn't improve. There's another parameter STALENESS which is currently set to UNUSABLE. I have been browsing web but couldn't find much that can help getting the mview of of unusable state.
Could you guys please answer some of my quesiton:
1) Is there a way to get the mview out of unusable state without having to drop it and rebuild?
2) Why would my Mview be running forever and not coming back with any success/failure message?
Note: This mview has dependencies on multiple dimension and fact tables.
Thanks.
Regards
Akshay

You might want to refer to the "Managing Materialized Views" chapter of the Oracle Tuning Guide, and the "Advanced Materialized Views" chapter in the data warehousing guide.
Are you creating materialized view logs and performing a fast (or forced) refresh? The logs can save considerable time as they record changes to the base data, and can be used for fast refreshes. A forced refresh will defer to a fast refresh when it can, or do a full refresh when it must.
Are you taking advantage of partitions?
Look for bottlenecks in your query using the tools already suggested.
You might need to drop and recreate the materialized view if its based on a prebuilt table that was never refreshed.

Refresh a materialized view in parallel

Hi,
To improve the refresh performance for a materialized view, I set up it to be refreshed in parallel. The view can be refreshed successfully, however, I did not see the view is refreshed in parallel from session browser, can someone let me know if I miss any steps?
1) In DB A (running 8 CPUs), set up the base table to be parallelized, the base table is called table1
ALTER TABLE A.table1 PARALLEL ( DEGREE Default INSTANCES Default );
2) In Database A, set up the materialized view log
CREATE MATERIALIZED VIEW LOG
ON table1 WITH primary key
INCLUDING NEW VALUES;
3) In Database B (in the same server with Database A), there is an existing table called table1, prebuilt with millions of records in the table. Due to the size of table1, I have to use prebuilt option
Drop MATERIALIZED VIEW table1;
CREATE MATERIALIZED VIEW table1 ON PREBUILT TABLE
REFRESH FAST with primary key
AS
select /*+ PARALLEL(table1, 8) */ *
from table1@A;
4) in Database B, I executed this stored procedure -
EXECUTE DBMS_MVIEW.REFRESH(LIST=>'table1',PARALLELISM=>8, METHOD=>'F');
Thanks in advance!
Liz

I checked the execution plan when I executed the refresh stored procedure again, Does the plan indicates the parallel execution is working or not?
SELECT STATEMENT REMOTE ALL_ROWSCost: 32,174 Bytes: 775,392 Cardinality: 1,182
13 PX COORDINATOR
12 PX SEND QC (RANDOM) PARALLEL_TO_SERIAL SYS.:TQ10001 :Q1001
     11 NESTED LOOPS PARALLEL_COMBINED_WITH_PARENT :Q1001
     9 NESTED LOOPS PARALLEL_COMBINED_WITH_PARENT :Q1001Cost: 32,174 Bytes: 775,392 Cardinality: 1,182
          6 BUFFER SORT PARALLEL_COMBINED_WITH_CHILD :Q1001
               5 PX RECEIVE PARALLEL_COMBINED_WITH_PARENT :Q1001
                    4 PX SEND ROUND-ROBIN PARALLEL_FROM_SERIAL SYS.:TQ10000 A
                         3 VIEW APP_USR. Cost: 12,986 Bytes: 6,070,196 Cardinality: 275,918
                              2 HASH UNIQUE Cost: 12,986 Bytes: 9,105,294 Cardinality: 275,918
                                   1 TABLE ACCESS FULL TABLE A.MLOG$_TABLE1 A Cost: 10,518 Bytes: 9,105,294 Cardinality: 275,918
     8 PARTITION RANGE ITERATOR PARALLEL_COMBINED_WITH_PARENT :Q1001Cost: 0 Cardinality: 1
     7 INDEX RANGE SCAN INDEX (UNIQUE) PARALLEL_COMBINED_WITH_PARENT A.PK_TABLE1 :Q1001Cost: 0 Cardinality: 1
10 TABLE ACCESS BY LOCAL INDEX ROWID TABLE PARALLEL_COMBINED_WITH_PARENT A.TABLE1 :Q1001Cost: 0 Bytes: 634 Cardinality: 1

Way to add an additional column that acts as a PK in a Materialized View?

I want to create a materialized view that also has an index column. A unique number sequence assigned to each record. Is this even possible? Can a materialized view do this, and would the indexes stay the same on a refresh?
Any help would be greatly appreciated.

Materialized views are copies of data or aggregations. They can be used to replicate all or part of a table, or to replicate the result of a query against multiple tables. The big advantage they have is that unlike a standard view, which can return the same data, you can index the materialized view to improve performance. These indexes will be maintained and updated during the refresh process.
Refer to Metalink document: 76673.1 or the Oracle Concepts manual for more information.

Problem encountered when join two remote tables in a materialized view

I'm using oracle 9.2.0.6
1> I have two tables:
CREATE TABLE TEST
A VARCHAR2(100 BYTE),
C DATE
CREATE TABLE TEST1
A VARCHAR2(100 BYTE),
B TIMESTAMP
2>. I defined a prebuild table:
CREATE TABLE MV_TEST1
ID1 ROWID,
A VARCHAR2(100 BYTE),
ID2 ROWID,
B TIMESTAMP(6),
C DATE
3> I created mview logs:
CREATE MATERIALIZED VIEW LOG ON PSI_TEST.TEST
WITH ROWID
INCLUDING NEW VALUES;
CREATE MATERIALIZED VIEW LOG ON PSI_TEST.TEST1
WITH ROWID
INCLUDING NEW VALUES;
4> when I create mview:
CREATE MATERIALIZED VIEW PSI_TEST.MV_TEST1
ON PREBUILT TABLE WITH REDUCED PRECISION
REFRESH FAST ON DEMAND
WITH PRIMARY KEY
AS
select
test.rowid id1,
test.a,
test1.rowid id2,
test1.b,
cast(null as date) c
from test , test1
where test.a = test1.a(+);
It is created successfully.
5> problem:
when I use remote tables to do the same thing, say test and test1 are in another instance and are connected by a dbLink, I couldn't create the mview successfully:
CREATE MATERIALIZED VIEW PSI_TEST.MV_TEST1
ON PREBUILT TABLE WITH REDUCED PRECISION
REFRESH FAST ON DEMAND
WITH PRIMARY KEY
AS
select
a.rowid id1,
a.a,
b.rowid id2,
b.b,
cast(null as date) c
from test@dbl a, test1@dbl b
where a.a = b.a(+);
when run above statement, I got:
ORA-12015: cannot create a fast refresh materialized view from a complex query
Any ideas? Or joining two table through a dblink for a mview is not allowed at all?
Thanks in advance.

No one has a clue?
Message was edited by:
lzhwxy

Creating Materialized View in Toad

As with every new endeavor - come new questions.
I am trying to create a materialized view via toad (first time doing this). Now - toad is great, since it provides a kind-of wizard interface. Hence, one doesn't have to completely code in the create statement, with all of the options, etc.
Instead, in toad, one can (via the schema browser), go into the Materialized Views tab and click on create new. This opens a window with 6 tabs: Basic Info, Refresh Info, Physical Attributes, Query, Partitions & Subpartition Template.
In the Basic Info, one can put a check mark next to some of the following options:
-Build Deffered
-Parallel
-Cache
-Logging
-Using index
-Allow updates
etc.
I have read that build deferred refers to whether or not you would like the view to be created automatically or to be deferred.
Anyways, I tried to find Toad documentation to explain each of the options in-detail. No success. Hence, am researching each part piece by piece.
But - here is my question:
In the Query tab, we are asked to specify a query. I am assuming that the query does not have to be a full query, beginning with CREATE MATERIALIZED VIEW mv_table1 REFRESH FAST etc...
Since in this 'wizard', we are providing the MV name at the top of the dialog box, we are checking the 'options' in the other tabs, etc.
And so, I assumed that the query should merely be a select query in the window mentioned above.
So I entered the following:
select * from table1@remote_db;
When I go to verify the syntax, I get the error message: ORA_00911: invalid character. It seems to be pointing to my db_link (remote_db). I have been using this link throughout many places, w/o any problems.
Has anyone created a MV in toad before? Any links to good toad documentation would be helpful as well.
Thanks.

(This is fun... ;))
For anyone endeavoring this in the future, below I have attached the prerequisites required in order to create a materialized view (can also be found @ http://download-uk.oracle.com/docs/cd/B14117_01/server.101/b10759/statements_6002.htm):
Prerequisites
The privileges required to create a materialized view should be granted directly rather than through a role.
To create a materialized view in your own schema:
You must have been granted the CREATE MATERIALIZED VIEW system privilege and either the CREATE TABLE or CREATE ANY TABLE system privilege.
You must also have access to any master tables of the materialized view that you do not own, either through a SELECT object privilege on each of the tables or through the SELECT ANY TABLE system privilege.
To create a materialized view in another user's schema:
You must have the CREATE ANY MATERIALIZED VIEW system privilege.
The owner of the materialized view must have the CREATE TABLE system privilege. The owner must also have access to any master tables of the materialized view that the schema owner does not own (for example, if the master tables are on a remote database) and to any materialized view logs defined on those master tables, either through a SELECT object privilege on each of the tables or through the SELECT ANY TABLE system privilege.
To create a refresh-on-commit materialized view (ON COMMIT REFRESH clause), in addition to the preceding privileges, you must have the ON COMMIT REFRESH object privilege on any master tables that you do not own or you must have the ON COMMIT REFRESH system privilege.
To create the materialized view with query rewrite enabled, in addition to the preceding privileges:
If the schema owner does not own the master tables, then the schema owner must have the GLOBAL QUERY REWRITE privilege or the QUERY REWRITE object privilege on each table outside the schema.
If you are defining the materialized view on a prebuilt container (ON PREBUILT TABLE clause), then you must have the SELECT privilege WITH GRANT OPTION on the container table.
The user whose schema contains the materialized view must have sufficient quota in the target tablespace to store the master table and index of the materialized view or must have the UNLIMITED TABLESPACE system privilege.
When you create a materialized view, Oracle Database creates one internal table and at least one index, and may create one view, all in the schema of the materialized view. Oracle Database uses these objects to maintain the materialized view data. You must have the privileges necessary to create these objects.

Need help on tuning materialized view refresh

Hi All,
I am working on materialized view refresh tuning.Initially it was complete refresh and used to take more than 90 mins to complete.
I changed it to fast refresh now it is completing fast. Now i have partitioned the base tables gl_balances and gl_code_combinations of column code_combination_id and created a local index on column code_combination_id then i am trying to partition the materialized on the same column to take advantage of partition change tracking.
Size of gl_balances base tables is 40Gb and all others tables sizes are small. In where clause there all the 4 tables are mapped. If i will create the partition only on code_combination_id will i the materialized will become the candidate for partition change tracking. As i know it will be applicable for PCT. I need expert advice on this.
While doing a fast refresh. the refresh takes less time. when there is a change in gl_balances , gl_code_combinations or gl_periods it completes in 20-30 mins. When there is a change in gl_set_of_books tables. It creates a problem here.DEL query takes more than 48 hours to complete.
CREATE MATERIALIZED VIEW apps.BAL_PART
REFRESH FAST ON DEMAND
ENABLE QUERY REWRITE as
SELECT GL.GL_CODE_COMBINATIONS21.ROWID C1,GL.GL_BALANCES21.ROWID C2, GL.GL_SETS_OF_BOOKS.ROWID C3,
GL.GL_PERIOD.ROWID C4,
"GL"."GL_BALANCES21"."ACTUAL_FLAG" ,
"GL"."GL_BALANCES21"."CURRENCY_CODE" ,
"GL"."GL_BALANCES21"."PERIOD_NUM" ,
"GL"."GL_BALANCES21"."PERIOD_YEAR" ,
"GL"."GL_BALANCES21"."SET_OF_BOOKS_ID" "SOB_ID",
"GL"."GL_CODE_COMBINATIONS21"."CODE_COMBINATION_ID" "CCID",
"GL"."GL_CODE_COMBINATIONS21"."SEGMENT1" ,
"GL"."GL_CODE_COMBINATIONS21"."SEGMENT10" ,
"GL"."GL_CODE_COMBINATIONS21"."SEGMENT11" ,
"GL"."GL_CODE_COMBINATIONS21"."SEGMENT12" ,
"GL"."GL_CODE_COMBINATIONS21"."SEGMENT13" ,
"GL"."GL_CODE_COMBINATIONS21"."SEGMENT14" ,
"GL"."GL_CODE_COMBINATIONS21"."SEGMENT2" ,
"GL"."GL_CODE_COMBINATIONS21"."SEGMENT3" ,
"GL"."GL_CODE_COMBINATIONS21"."SEGMENT4" ,
"GL"."GL_CODE_COMBINATIONS21"."SEGMENT5" ,
"GL"."GL_CODE_COMBINATIONS21"."SEGMENT6" ,
"GL"."GL_CODE_COMBINATIONS21"."SEGMENT7" ,
"GL"."GL_CODE_COMBINATIONS21"."SEGMENT8" ,
"GL"."GL_CODE_COMBINATIONS21"."SEGMENT9" ,
"GL"."GL_PERIODS"."PERIOD_NAME" ,
NVL("GL"."GL_BALANCES21"."BEGIN_BALANCE_CR", 0) Open_Bal_Cr,
NVL("GL"."GL_BALANCES21"."BEGIN_BALANCE_CR", 0) +
NVL("GL"."GL_BALANCES21"."PERIOD_NET_CR", 0) Close_Bal_Cr,
NVL("GL"."GL_BALANCES21"."BEGIN_BALANCE_DR", 0) Open_Bal_Dr,
NVL("GL"."GL_BALANCES21"."BEGIN_BALANCE_DR", 0) +
NVL("GL"."GL_BALANCES21"."PERIOD_NET_DR", 0) Close_Bal_Dr,
NVL("GL"."GL_BALANCES21"."BEGIN_BALANCE_DR", 0) -
NVL("GL"."GL_BALANCES21"."BEGIN_BALANCE_CR", 0) Open_Bal,
NVL("GL"."GL_BALANCES21"."BEGIN_BALANCE_DR", 0) -
NVL("GL"."GL_BALANCES21"."BEGIN_BALANCE_CR", 0) +
NVL("GL"."GL_BALANCES21"."PERIOD_NET_DR", 0) -
NVL("GL"."GL_BALANCES21"."PERIOD_NET_CR", 0) Close_Bal,
NVL("GL"."GL_BALANCES21"."PERIOD_NET_CR", 0) Period_Cr,
NVL("GL"."GL_BALANCES21"."PERIOD_NET_DR", 0) Period_Dr
FROM GL.GL_CODE_COMBINATIONS21,
GL.GL_BALANCES21,
GL.GL_SETS_OF_BOOKS,
GL.GL_PERIODS
WHERE GL.GL_BALANCES21.CODE_COMBINATION_ID =GL.GL_CODE_COMBINATIONS21.CODE_COMBINATION_ID
AND GL.GL_SETS_OF_BOOKS.SET_OF_BOOKS_ID = GL.GL_BALANCES21.SET_OF_BOOKS_ID
AND GL.GL_PERIODS.PERIOD_NUM = GL.GL_BALANCES21.PERIOD_NUM
AND GL.GL_PERIODS.PERIOD_YEAR = GL.GL_BALANCES21.PERIOD_YEAR
AND GL.GL_PERIODS.PERIOD_TYPE = GL.GL_BALANCES21.PERIOD_TYPE
AND GL.GL_PERIODS.PERIOD_NAME = GL.GL_BALANCES21.PERIOD_NAME
AND GL.GL_PERIODS.PERIOD_SET_NAME = GL.GL_SETS_OF_BOOKS.PERIOD_SET_NAME
and gl.GL_CODE_COMBINATIONS21.summary_flag != 'Y'TRACE 1046 del statement
DELETE FROM "APPS"."apps.BAL_PART" SNA$
WHERE "C3" IN (SELECT /*+ NO_MERGE */ * FROM (SELECT
CHARTOROWID("MAS$"."M_ROW$$") RID$     FROM "GL"."MLOG$_GL_SETS_OF_BOOKS"
"MAS$"   WHERE "MAS$".SNAPTIME$$ > :B_ST1 ) AS OF SNAPSHOT(:B_SCN) MAS$)
call     count       cpu    elapsed       disk      query    current        rows
Parse        1      0.00       0.00          0          0          0           0
Execute      1 17759.00 171782.99 159422121    1267371 2564144739           0
Fetch        0      0.00       0.00          0          0          0           0
total        2 17759.00 171782.99 159422121    1267371 2564144739           0
Misses in library cache during parse: 1
Misses in library cache during execute: 1
Optimizer mode: ALL_ROWS
Parsing user id: 175 (APPS)   (recursive depth: 1)
Rows     Row Source Operation
      0 DELETE apps.BAL_PART (cr=0 pr=0 pw=0 time=0 us)
193128740   NESTED LOOPS (cr=592437 pr=592422 pw=0 time=945244160 us cost=339302 size=168 card=1)
      3    SORT UNIQUE (cr=7 pr=0 pw=0 time=15832 us cost=2 size=138 card=1)
     24     TABLE ACCESS FULL MLOG$_GL_SETS_OF_BOOKS (cr=7 pr=0 pw=0 time=19 us cost=2 size=138 card=1)
193128740    INDEX RANGE SCAN C3BOOKS (cr=592430 pr=592422 pw=0 time=789499200 us cost=339299 size=3318314250 card=110610475)(object id 2114736)
error during execute of EXPLAIN PLAN statement
ORA-08187: snapshot expression not allowed here
parse error offset: 314
Elapsed times include waiting on following events:
Event waited on                             Times   Max. Wait Total Waited
----------------------------------------   Waited ---------- ------------
db file sequential read                   159520897        2.12     144415.96
latch: cache buffers chains                   134        0.06          0.68
latch: undo global data                        33        0.02          0.15
latch: object queue header operation          521        0.02          0.53
log file switch (private strand flush incomplete)
                                                532        0.31         28.26
resmgr:cpu quantum                            155        1.40         13.49
resmgr:internal state change                   25        0.11          2.21
latch free                                     10        0.00          0.00
latch: cache buffers lru chain                  4        0.00          0.00
rdbms ipc reply                               489        0.02          0.54
reliable message                              587        0.00          0.56
latch: row cache objects                        3        0.00          0.00
********************************************************************************GL_SETS_OF_BOOKS has only 6 rows. I know there is complete refresh as a option which will again take more than 90 mins.
I want to do the fast refresh. Tables rows details below.
SQL> select count(*) from gl.gl_code_combinations21;
COUNT(*)
3075255
SQL> select count(*) from gl.GL_PERIODS;
COUNT(*)
1160
SQL> select count(*) from gl.gl_balances21;
COUNT(*)
477613527
SQL> select count(*) from gl.gl_sets_of_books;
COUNT(*)
6gl_sets_of_books has less rows. Whenever there is a change then it mapped to huge rows hence during materialized view has delete huge number of rows.
select count(*) from apps.BAL_PART group by C3;
C3 is the rowid which is present in create materialized statement.
COUNT(*)
292927011
210215
69330
184406971
Is there any way to improve the plan. As i created a partition on code_combination_id and local index on code_combination_id which will not help in set_of_books_id case. I dont PCT will help here or not. Is it possible to use PCT refresh by equipartitioning only one column in where clause.
Please assist me in improving refresh of materialized view using fast refresh.
Thanks and Regards,
Edited by: user646034 on Feb 23, 2013 11:13 PM
Edited by: user646034 on Feb 23, 2013 11:19 PM
Edited by: user646034 on Feb 23, 2013 11:46 PM
Edited by: user646034 on Feb 25, 2013 11:46 AM

Hi
The below explain without index and with index.
/* MV_REFRESH (DEL) */ DELETE FROM "APPS"."BAL_PART
" SNA$ WHERE "C3" IN (SELECT /*+ NO_MERGE */ * FROM (SELECT
CHARTOROWID("MAS$"."M_ROW$$") RID$     FROM
"GL"."MLOG$_GL_SETS_OF_BOOKS" "MAS$"   WHERE "MAS$".SNAPTIME$$ > :B_ST1
) AS OF SNAPSHOT(:B_SCN) MAS$)
Plan hash value: 2704021294
                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                       | Id | Operation            | Name                          | E-Rows |E-Bytes| Cost (%CPU)| E-Time   |
|   0 | DELETE STATEMENT     |                               |        |       |   339K(100)|          |
|   1 | DELETE              | BAL_PART                          |        |       |            |          |
|   2 |   NESTED LOOPS       |                               |      1 |   168 |   339K (1)|999:59:59 |
|   3 |    SORT UNIQUE       |                               |      1 |   138 |     2   (0)| 00:02:31 |
|   4 |     TABLE ACCESS FULL| MLOG$_GL_SETS_OF_BOOKS        |      1 |   138 |     2   (0)| 00:02:31 |
|   5 |    INDEX RANGE SCAN | C3BOOKS                       |    110M| 3164M|   339K (0)|999:59:59 |
If i will not use the C3 index then the query will use the belolw plan, I guess this will also take same time or more time.
|   0 | DELETE STATEMENT       |                               |        |       | 9743K(100)|          |
|   1 | DELETE                | BAL_PART                          |       |            |          |

A quicker way to create a materialized view?

Hi,
I'm new to replication and have been reading doco on it. Basically I want to set up read-only materialized views in oracle 11g.
So on the replicated database I created a db link that points to the master db.
Then I created a materialized view that does "select * from table@dblink" and that is forced refresh every 10 minutes.
Problem is that the table on the master db is large and the network link is slow, so the "create materialized view" statement would take a long time to complete.
So I think to speed up the process I would export/import the table from master to replicated db, then make this imported table a materialized view somehow.
Is this possible, and if so, how?
Thanks.
Long

If you are able to use other means (exp-imp, expdp/impdp, unload-sqlldr, externalfile etc) to copy the data faster (as a normal table in the target database), you can then use the "ON PREBUILT TABLE" clause in the CREATE MATERAILIZED VIEW statement.
See
http://download.oracle.com/docs/cd/B19306_01/server.102/b14200/statements_6002.htm#i2063793
Alternatively, you must first verify if the CREATE is slow only because of SQL Query performance on the source, rather than because of network bandwidth/latency. If query execution needs to be tuned, that would be a cleaner solution, because the Refresh can then be automatically handled by Oracle.
Remember that you can use Materialized View Log(s) on the source tables to enable FAST Refresh's as well.

Trigger on a materialized view

i have a materialized view on a prebuilt table..
is it possible to write after triggers on the table?
regards
raj

check a last updated column in the materialised view after a refresh..
and collect those rows and process those as :new rows....
but i need to keep on checking if the materialized view is refreshed or not.....You would need to use a job to repeatedly check the MVIEW for any rows which have been updated since the last time the job ran. You would need some form of log to record the time your job used to make that check. It is a bit clunky, but if you read Tom Kyte's article you'll understand why triggers don't work.
An alternative approach would be to have the source tables which feed the MVIEW also trigger whatever additional processing it is that you require. But this would work only in a handful of situations.
Is this related to other question about Java Stored Procedures?
Cheers, APC
blog: http://radiofreetooting.blogspot.com

How to create a Complex Organization Index Materialized View Example

Hi
I have a 11g database that I'm trying to create a complex Materialized View that I would like to make Organization Index? How do I specify what I want for a primary Key?
CREATE MATERIALIZED VIEW RCS_STG.MV_NEXT_HOP_iot
ORGANIZATION INDEX
AS
SELECT r2.resource_key, r1.resource_key resource_key2, r2.resource_full_path_name, device_name, device_model,
service_telephone_number, service_package_name, telephone_number.telephone_number_key, c1.created_on
FROM network_resource PARTITION (network_resource_subinterface) r1,
connection c1,
network_resource PARTITION (network_resource_subinterface) r2,
device d1,
tn_network_resource,
telephone_number
WHERE r1.resource_key = c1.resource1_key
AND c1.resource2_key = r2.resource_key
AND d1.device_key = r2.device_key
AND tn_network_resource.resource_key(+) = r2.resource_key
AND telephone_number.telephone_number_key(+) = tn_network_resource.telephone_number_key
UNION ALL
SELECT r1.resource_key, r2.resource_key resource_key2, r1.resource_full_path_name, device_name, device_model,
service_telephone_number, service_package_name, telephone_number.telephone_number_key, c1.created_on
FROM network_resource PARTITION (network_resource_subinterface) r1,
connection c1,
network_resource PARTITION (network_resource_subinterface) r2,
device d1,
tn_network_resource,
telephone_number
WHERE r1.resource_key = c1.resource1_key
AND c1.resource2_key = r2.resource_key
AND d1.device_key = r1.device_key
AND tn_network_resource.resource_key(+) = r1.resource_key
AND telephone_number.telephone_number_key(+) = tn_network_resource.telephone_number_key
I get an error message ORA-25175: no PRIMARY KEY constraint found
I would like to specify resource_key, resource_key2, and service_telephone_number as my primary key?

Ah,
I get it now. This is what I did.
CREATE TABLE mv_next_hop_iot
resource_key NUMBER (38),
resource_key2 NUMBER (38),
resource_full_path_name VARCHAR2 (256 BYTE),
device_name VARCHAR2 (64 BYTE),
device_model VARCHAR2 (64 BYTE),
service_telephone_number VARCHAR2 (20 BYTE),
service_package_name VARCHAR2 (64 BYTE),
telephone_number_key NUMBER (38),
created_on DATE,
CONSTRAINT mv_next_hop_pk PRIMARY KEY (resource_key, resource_key2, service_telephone_number)
ORGANIZATION INDEX
CREATE MATERIALIZED VIEW rcs_stg.mv_next_hop_iot
ON PREBUILT TABLE
AS
/* Formatted on 2010/06/10 1:39:04 PM (QP5 v5.149.1003.31008) */
SELECT resource_key, resource_key2, resource_full_path_name, device_name, device_model, service_telephone_number,
service_package_name, telephone_number_key, created_on
FROM (SELECT r2.resource_key, r1.resource_key resource_key2, r2.resource_full_path_name, device_name, device_model,
NVL (service_telephone_number, ' ') AS service_telephone_number, service_package_name,
telephone_number.telephone_number_key, c1.created_on
FROM network_resource PARTITION (network_resource_subinterface) r1,
connection c1,
network_resource PARTITION (network_resource_subinterface) r2,
device d1,
tn_network_resource,
telephone_number
WHERE r1.resource_key = c1.resource1_key
AND c1.resource2_key = r2.resource_key
AND d1.device_key = r2.device_key
AND tn_network_resource.resource_key(+) = r2.resource_key
AND telephone_number.telephone_number_key(+) = tn_network_resource.telephone_number_key
UNION ALL
SELECT r1.resource_key, r2.resource_key resource_key2, r1.resource_full_path_name, device_name, device_model,
NVL (service_telephone_number, ' ') AS service_telephone_number, service_package_name,
telephone_number.telephone_number_key, c1.created_on
FROM network_resource PARTITION (network_resource_subinterface) r1,
connection c1,
network_resource PARTITION (network_resource_subinterface) r2,
device d1,
tn_network_resource,
telephone_number
WHERE r1.resource_key = c1.resource1_key
AND c1.resource2_key = r2.resource_key
AND d1.device_key = r1.device_key
AND tn_network_resource.resource_key(+) = r1.resource_key
AND telephone_number.telephone_number_key(+) = tn_network_resource.telephone_number_key)
Many thanks. the PREBUILT TABLE is the secret.

Inserting into a base table of a materialized view takes lot of time.......

Dear All,
I have created a materialized view which refreshes on commit.materialized view is enabled query rewrite.I have created a materialized view log on the base table also While inserting into the base table it takes lot of time................Can u please tell me why?

Dear Rahul,
Here is my materialized view..........
create materialized view mv_test on prebuilt table refresh force on commit
enable query rewrite as
SELECT P.PID,
SUM(HH_REGD) AS HH_REGD,
SUM(INPRO_WORKS) AS INPRO_WORKS,
SUM(COMP_WORKS) AS COMP_WORKS,
SUM(SKILL_WAGE) AS SKILL_WAGE,
SUM(UN_SKILL_WAGE) AS UN_SKILL_WAGE,
SUM(WAGE_ADVANCE) AS WAGE_ADVANCE,
SUM(MAT_AMT) AS MAT_AMT,
SUM(DAYS) AS DAYS,
P.INYYYYMM,P.FIN_YEAR
FROM PROG_MONTHLY P
WHERE SUBSTR(PID,5,2)<>'PP'
GROUP BY PID,P.INYYYYMM,P.FIN_YEAR;
Please help me if query enable rewrite does any performance degradation......
Thanks & Regards
Kris

Advantage of ON PREBUILT in Materialized views

Similar Messages

Maybe you are looking for