OWB Dimensions

DBME:Oracle 10gR2
OS:WindowsXp
hello,
I am using OWB 10gR2 .I am working with dimensions.I know the basics of DWH but while reading "10R2 owb user guide" I get confused with some concepts.
Can any one send me some good link from where I can read Dimension using OWB 10gR2. That might include detail.
thanks
tanveer

Hi,
I hope this link would help :
http://download-uk.oracle.com/docs/cd/B31080_01/doc/owb.102/b28223/ref_dim_objects.htm

Similar Messages

CBO: OWB Dimension Performance Isssue (DIMENSION_KEY=DIM_LEVEL_ID)

Hi
In my opinion the OWB Dimensions are very useful, but sometimes there are some Performance Issues.
I am working with the OWB Dimensions quite a lot and with the big Dimensions ( > 100.000 rows) , i often get some Performance problems when OWB generates the code to Load (Merge Step) or Lookup these Dimensions.
OWB Dimensions have a PK on DIMENSION_KEY and Level Surrogate IDs which are equal to the DIMENSION_KEY if the The Row is an Element of that Level (and not a Parent Hierarchic Element)
I am hunting the Problem down to the Condition DIMENSION_KEY= (DETAIL_)LEVEL_SURROGATE_ID. The OWB does that to get only the Rows with (Detail-) Level Attributes.
But it seems, that the CBO isn´t able to predicted the Cardinality right. The CBO always assume, that the Result Cardinality of that Condition is 1 row. So I assume that Conditon is the reason for the "bad" Execution Plans, the Execution Plan
"NESTED LOOPS OUTER" With the Inline View with Cardinality = 1;
Example:
SELECT COUNT(*) FROM DIM_KONTO_TAB WHERE DIMENSION_KEY= KONTO_ID;
--2506194
Explain Plan for:
SELECT DIMENSION_KEY, KONTO_ID
FROM DIM_KONTO_TAB where DIMENSION_KEY= KONTO_ID;
+| Id | Operation | Name | Rows | Bytes | Cost (%CPU)| Time | Pstart| Pstop |+
+| 0 | SELECT STATEMENT | | 1 | 12 | 12568 (3)| 00:00:01 | | |+
+| 1 | PARTITION HASH ALL | | 1 | 12 | 12568 (3)| 00:00:01 | 1 | 8 |+
+|* 2 | TABLE ACCESS STORAGE FULL| DIM_KONTO_TAB | 1 | 12 | 12568 (3)| 00:00:01 | 1 | 8 |+
Predicate Information (identified by operation id):
+2 - STORAGE("DIMENSION_KEY"="KONTO_ID")+
filter("DIMENSION_KEY"="KONTO_ID")
Or: For Loading an SCD2 Dimension:
+|* 12 | FILTER | | | | | | Q1,01 | PCWC | |+
+| 13 | NESTED LOOPS OUTER | | 328K| 3792M| 3968 (2)| 00:00:01 | Q1,01 | PCWP | |+
+| 14 | PX BLOCK ITERATOR | | | | | | Q1,01 | PCWC | |+
+| 15 | TABLE ACCESS STORAGE FULL | OWB$KONTO_STG_D35414 | 328K| 2136M| 27 (4)| 00:00:01 | Q1,01 | PCWP | |+
+| 16 | VIEW | | 1 | 5294 | | | Q1,01 | PCWP | |+
+|* 17 | TABLE ACCESS STORAGE FULL | DIM_KONTO_TAB | 1 | 247 | 489 (2)| 00:00:01 | Q1,01 | PCWP | |+
I tried a lot:
- statistiks are gathered often, with monitoring Informations and (Frequencey-)Histograms and the Conditions Colums
- created extend Statistiks DBMS_STATS.CREATE_EXTENDED_STATS(USER, 'DIM_KONTO_TAB', '(DIMENSION_KEY, KONTO_ID)')
- created combined idx one DIMENSION_KEY, LEVEL_SURROGATE_ID
- red a lot
- hinted the Querys in OWB ( but it seems the inline View is to complex to use a Hash Join)
Next Step:
-Tracing the Optimizer CBO Events.
Does some one has an Idea how-to help the CBO to get the cardinality right?
If you need more Information, please tell me.
Thanks a lot.
Moritz

Hi Patrick,
For a relational dimension, these values must be unique within the LEVEL. It is not required to be a numeric ID (although that follows the best practices of surrogate keys best).
If you use the same sequence for the dimension you have insured that each entry in the entire dimension is unique. Which means that you can move your data as is into OLAP solutions. We will do this as well in the next major release.
Hope that helps,
Jean-Pierre

10.2 OWB Dimension "binding" erases all Table metadata!

I need some help with dimension definitions in OWB. I have a dimension that I created and deployed as ROLAP which includes its corresponding table.
I added a number of indexes on the table and some default values on the columns - trying to avoid NULLS in the dimension fields.
I then realized I needed one additional attribute in the dimension. I put the additional attribute in, selected "Bind" and poof - no more indexes, constraints or default values. It appears to have completely recreated the table metadata.
Also, the column ordering of the corresponding table for the dimension appears to be random.
Is there some better way to control this? I'm sure I'll have to add future attributles to existing dimenions!
TIA,
Mike

OK, after perusing the manual for awhile, I found that to add columns to an existing dimension I need to:
1. Go into dimension object editor, select the Storage tab and set it to "Manual" instead of Star.
2. Add the columns to both the dimension and table
3. right click the dimension to show "detailed view"
4. manually map the column
5. deploy (after juggling the creation type of the dimension and table)
I don't find it at all intuitive and if you forget you wipe out your table metadata for the dimension but at least it works.

OWB Dimension Key

Hi Everyone,
I have a simple question for you guys, can anyone explain to me why that my PK from owb on Type I SCD is always starting a the number of rows that I inserts? If I'm loading a dimension for the first time, beit INSERT/UPDATE or UPDATE/INSERT and I'm loading n rows, the sequence starts at n instead of 1, I know that I should bother with that since it is only a PK but would like to understand what is being done in the background
Thank you
Jacques

Hi
OWB create the sequence to start at 1, when deployed it is ok, but when i run my mapping, it is like it is caching all the key for all rows and then generate new ones when inserting, I did start my sequence at minus n (-n) and then it started at 1, on the insert. Just curious to know what is happening under the hood.
Thanks for your reply

Dimension out of synch with relational table

Hi Gurus,
We have an issue of Dimension Loading in OWB10grel2.
My Dimension DPROD Details are as below..
Storage :star
SCD type:1
Levels : Lev1,2,3...15 with each level having ID,CD,DESCR as level attributes.
Hierarchy: prodlev1,prodlev2....prodkey
All the attributes,heirarchies,levels are defined properly.
I have a mapping to populate this dimension
RELATIONALPROD---->DPROD.
The RELATIONALPROD is the staging table which is equivalent to Dimension table(except the dim keys) and acts as a source for populating DPROD.
we have loaded dimension succesfully and refreshing it on a need basis.
However;we are now facing a strange issue of some attributes in few records are being out of synch in DPROD.
For e.g the Level2 DESCR,Level2CD in DPROD for a given prod_key are not the same as in
RELATIONALPROD table for that prod_key.
When we have tried re-executing the mapping it doesnot update the DPROD with the values in RELATIONALPROD for that matching prod_key.
Anyone faced the same issue?
Any idea on why this happens and could you please suggest any workarounds/methods to overcome this problem.
Thanks in advance!
Best Regards,
Sridh

Hi David,
I am a bit confused..Can you please give me more details please
Futhur,I have the following scenario
I have the below defined as attributes in the OWB dimension
ID surrogate
CD Business
DESCR
x
y
z
And my levels for level10 till level50 has the following three as 'Dimension attribute names' with 'Applicable' option checked
ID
CD
DSCR
and the top most level with the below as 'Dimension attribute names' with 'Applicable' option checked
ID
CD
DESCR
x
y
z
A mapping from source to dimension is created.
The problem is the mapping doesnot update CD information at the levels in dimension based on the values in source.
Best Regards,
Sridhar

Time Dimension - holiday flag

Hi,
Does anyone have good idea how to implement holiday_flag in time dimension?
I wondered if I added "holiday_flag" attribute,
however, our universe is global and holiday date is different between each countries.
Some ideas has come up as below, but I don't think those are good idea.
1. To create holiday_flag variable which is dimensioned by time and country.
2. To create holiday_flag_XXX attribute for each country.
3. To create holiday function which returns flag which indicates holiday or not.( to do that, kind of holiday cube would be required?)
The holiday_flag would be used to count business day or check for NAFILL.
Any help would be appreciated.
Thanks in advance,
Yukari.

Thanks for your reply. Did you use the dimension wizard of OWB to create pure relational dimensions (Not owb/oracle dimensional structures) ?
How did you work out week, financial years pieces upon using this OWB dimension wizard?
Thanks for your help.

Import existing dimensions

Does anybody know of a way to import existing tables from a database into an OWB dimension? We have an existing datawarehouse, I am trying to re-use the time dimension that is already created and in use, but do not see a way to import a dimension.

Hi,
At present you cannot import a dimension in OWB.
Cheers, Patrick

Time Dimension Average @ Lev2

Hi,I'm having a problem with calculating the TB Average at Lev2 of my Time Dim.Time Dim: Week (Lev0), Month (Lev1), Year (Lev2), Time (Lev3)Week to Month OkMonth to Year Not OkMonth to Year calculates the average using the Month TB Average (average of average).I've tried using a Member calc [@AVG (SKIPBOTH,@LEVMBRS ("Time",0))] for the Year Member so it calculates the average based on the Lev0 figures.Don't really want to have a separate the Years into a different dimension.Thanks in advance.Mark

Thanks for your reply. Did you use the dimension wizard of OWB to create pure relational dimensions (Not owb/oracle dimensional structures) ?
How did you work out week, financial years pieces upon using this OWB dimension wizard?
Thanks for your help.

Time_Dimension_specifics

Hi people
I want to create a cube in olap analytic workspace 11g, I have a problem with time dimension. I have followed the instructions from here
http://st-curriculum.oracle.com/obe/db/11g/r1/olap/cube/buildicubes.htm. I tried to create time dimension having type time by hand , but the program just crashed.
and we needed database restart to kill a process, it hanged during compilation, when I tried 'Maintain Cube Cube_Name ' .
Also I have tried to use olaptrain's template TIME.XML, it compiled correctly , but ended showing the info in 'view data Cube_Name' , even when I don't choose the time dimension in Measure Data Viewer.

Thanks for your reply. Did you use the dimension wizard of OWB to create pure relational dimensions (Not owb/oracle dimensional structures) ?
How did you work out week, financial years pieces upon using this OWB dimension wizard?
Thanks for your help.

Discoverer error: VLD-0005

when I deploy the The Business Area (derived items from owb: dimensions and cube) in control center, I got the error: VLD-0005 oracle.wh.iobuilder.service.discoexpr.TokenMgrError:
Lexical error at line 1, column 1. Encountered: "\u6ed1" (28369), after : "" . I follow the steps according the Lessons on oracle website:
Business Intelligence Modeling.
anyone can help me? thank you?
Edited by: user10979657 on 2010-3-30 上午6:54

Can always try deploying to file and importing the EEX file from Discoverer administrator?
Cheers
David

Star Schema and Oracle 11gR2 ?

Star Schema and Oracle 11gR2 ?
I know the star schema (ROLAP) and implemented couple of them. Apart from general design principle of dimension, FACT, surrogate key etc, what are the specific items needed in Oracle 11gR2?
Some one talked about over 10 conditions/pre-requisits for Star Schema (ROLAP) implementations in Oracle 11gR2. I did some search, but I did not get any hits.
Do we design Star schema (ROLAP) differently in Oracle 11gR2?
Any pointer welcome.
Thanks in helping.

Hi,
from my experience there are no specific requirements for the star schema design when using owb 11.2.
When using the OWB ETL Option (extra license required), one may use the owb dimensions and cubes.
These make mapping development easier, since support for SCD2 is built into the dimension operators. Loading the cube is simplified because the lookup of the surrogate key from the dimension is built into the cube operator.
These owb objects will deploy specific dimension and fact tables. If you already have existing ones, you must modify them manually.
I implemented several projects without these advanced features. Baiscally I did the same in OWB what I would have done using hand-coded SQL and PL/SQL. And it worked just fine.
If you find those 10 conditions, please post them here. I'm curious to learn about them!
Regards,
Carsten.

About Surrogate Key and Dimension Key on OWB 10.2

Hi, everyone.
I am using OWB 10.2 and I have a question about Surrogate key and Dimension Key.
I indicated the foreign key as VARCHAR2 type in Fact Table and Dimension Key as VARCHAR2 type is operated as Primary key in Dimension Table. I made Single Level in Dimension Table.
I know that Dimension Key stores the surrogate ID for dimension and is the primary key of the table. Also, Surrogate ID should be only NUMBER type.
So, in this case, Surrogate ID is NUMBER type
Dimension key should be NUMBER type to store the surrogate ID.
But, Dimension key also should operate the primary to relate Foreign key as VARCHAR2 type.
How I can solve this confusing condition?
Please let me know that.
JWS

Hi JWS,
From a SQL point of view it should not be a problem to join a NUMBER field to a VARCHAR2 field because during execution there will be an implicite cast for the NUMBER value to a VARCHAR2 value. See the example below.
SELECT * FROM DUAL
WHERE 1 = '1'From an OWB point of view it is not possible to have a Dimension with an NUMBER value Key that has a relation to a VARCHAR2 value Foreign key in a Fact table. This is caused due to the creation of a Fact table in OWB in which the Foreign keys in it are build from de Dimension tables that refer to them.
You will loose the reference to the Dimension when changing the type of the Foreign Key.
To resolve this issue I would advise you to use a Sequence that generates your Surrogate Key (NUMBER type) for the Dimension table and store it in the Primary Key Column (VARCHAR2 type).
When validating the mapping you will get a warning, but when executing this should give no problems.
Regards,
Ilona

Help with Skip Level dimensions in OWB 10.2 (Paris)

Hi all,
I'm trying to understand how to set up and use a skip level dimension in OWB 10.2. A simplified version of my hierarchy would be:
Plant - Equipment - Assembly - Item
Where Some Items connect directly to equipment and others connect to assemblies. How sould I define this in the Dimension wizard or editor in OWB?
Can I create a single dimension hierarchy and in the Item level put Skip Level as "Equipment"?
OR
Do I need 2 hierarchies - one Plant-Equipment-Assembly-Item and the other Plant-Equipment-Item?
There is an example in the Users guide that created 2 hierarchies -
TOTAL-GROUP-PRODUCT and TOTAL-PRODUCT
In the example, they show the PRODUCT in TOTAL-PRODUCT setting the skip level field. I don't understand why that is necessary if it is 2 seperate hierarchies.
Any help/guidance would be appreciated!
Thanks,
Mike

Hi Donna,
Perhaps JP can expand on his answer. In my case, I did go the route of seperate hierarchies. It is a bit of work defining multiple hierarchies and my example was much simplified. We actually have about 8 levels in the primary hierarchy and 6 levels in the second. It does work well with the cubes by using 2 roles for the dimension - one for each hierarchy.
I saw that truly "ragged hierarchies" or "level hierarchies" can be defined in OWB and deployed as MOLAP but MOLAP has quite a few drawbacks - no SCDs and requires analytic workspaces etc. I haven't done anything with these yet.
Our data is a classic manufacturing Bill of Material so the level hierarchy would be the easiest to manage, though I'm not sure how easy it would be to report on! I do want to shield my end-users from CONNECT-BY madness.

How to handle parent-child dimension in OWB?????

i have a dimension have many levels,and the amount of levels is varing,so i cannot use the wizard to define my dimension ,seems with OWB i can only define dimension with certain amount of levels.......
my dimension data is stored with this format:
child parent
Los Angles US
US WORLD
it is called a parent-child relation ,how can i define dimension with parent-child relation witn OWB...
Can OWB do it ?????

i have a dimension have many levels,and the amount of levels is varing,so i cannot use the wizard to define my dimension ,seems with OWB i can only define dimension with certain amount of levels.......
my dimension data is stored with this format:
child parent
Los Angles US
US WORLD
it is called a parent-child relation ,how can i define dimension with parent-child relation witn OWB...
Can OWB do it ????? You must define a dimension, define the dimension levels and the hierarchy inside the dimension (a hierarchy can have as many levels as you want, parent-child relationship is a normal hierarchy concept - no rocket science here). Then you have to map the data source for the dimension levels appropriately. Please refer to the user manual, chapter 4 ("Defining dimensianal targets") for details.
Regards:
Igor

Role Dimensions in OWB

hi,
I have a fact table linked with the same dimension table more than once. The fact got validated properly, but when trying to execute the mapping ,values are not getting populated in the fact table. I came to know that such dimensions are known are role playing dimensions and I would like to know how can this be implemented in OWB . My Project is currently using OWB (9i) and is planning to migrate to 10g R2. Can somebody plz help me? I am stuck with two of my mappings due to this issue. Plz help
thanks in advance..

Does anyone know if the bug with roles has been fixed in 11, I suspect not?
I'm doing some testing with OWB Client 11.1.0.6.0 and Server 11.1.0.1.1 and I'm hitting problems when deploying a Cube.
ORA-06510: PL/SQL: unhandled user-defined exception
ORA-06512: at "OLAPSYS.CWM2_OLAP_CUBE", line 99
ORA-06512: at line 2
The cube has 14 dimensions, 7 of which are roles (different dimensions). If I remove one of the dimensions that has 3 roles and replace it with 3 normal dimensions then the cube deploys ok.
Cheers
Si

OWB Dimensions

Similar Messages

Maybe you are looking for