Bmm Model

Hi ,
I have a requirement like below.
I have three DIM and two fact.
Ex: D1,D2,D3 - F1,F2
Mapping like below.
D1--F1
D2--F2
D3-F1,F2
How to design bmm layer plz suggest me.
Thanks..

Please see the following rough sketch:-
http://i49.tinypic.com/2v3fnk4.png
The coloured circles represent suggested subject areas for the presentation layer.
Please mark if helpful / correct,
Andy.

Similar Messages

Error In BMM model

Hi Everyone,
I am getting a strange problem. When checking consistency of a fairly large BMM model named core - I am getting following error.
ERRORS:
Business Model Core:
[nQSError: 93001] Can not load library, oracore11.dll, due to, The specified module could not be found..
The specified module could not be found.
[nQSError: 15001] Could not load navigation space for subject area Core.
Now nobody else is getting this error. And if I just save the rpd ignoring this error - the reports run fine. However the error still persists and irritates.
I am using Oracle BI Admin tool Version 11.1.1.6.10. Repository Version 320.
Others are on the same version still they do not get it. Any thought please? It is killing me.
Thanks in advance,
Avishek

Just to update.. it is finally resolved.
Issue was oracle 10g r2 client was installed in my machine. I downloaded oracle 11g r2 and started installer - selected custom - then OCI and did the installation.
Once done, did the consistency check again and it worked fine.
Bizarre OBIEE! as it was working perfect till somebody added a column where he did some semi-complex calc and from that point onwards - it started refusing to accept the rpd consistent in my m/c.
Regards, Avishek

BMM model for subquery

Hi All,
I have a requirement to pull the generate sql for the report.
select distinct T.Sum,A.Region,A.Section
from A
join ( select sum(Amount),Region from B
group by Region)T
B on A.Region=T.Region
What is the best practice to generate this query with sub query in BMM layer?
Do we need to create a view ( select sum(Amount),Region from B
group by Region)with subquery in physical layer and join with Table A
or there is other better way of modelling in BMM layer.
Thanks,
Surya

Hi All,
Thanks for your kind reply. The problem is that there are 3-4 layers of nested queries. How can we tackle these issue on BMM layer?
for example
select distinct X.Amount,C.Amount2,C.Region,C.Section
select distinct T.Sum,A.Region,A.Section
from A
join ( select sum(Amount),Region from B
group by Region)T
B on A.Region=T.Region)X
on X.Region=C.Region and X.Section =C.Section
group by
C.Amount2,C.Region,C.Section
Thanks,
Surya

BMM model problem

Hi experts,
In Physical Layer, I defined 4 tables :
- Product Category
- Product Category Fact
- Product
- Product Fact
I drag and drop all the 4 tables from Physical Layer to Business Model and Mapping Layer. Then I build complex joins among the tables as below:
1) Product Category (Dim) -> Product (Dim) -> Product Fact (Fact)
2) Product Category (Dim) -> Product Category Fact (Fact)
Then I further drag and drop all the 4 tables from BMM Layer to Presentation Layer. Then I use OBIEE Answer to start query.
However , I got error when querying columns from Product Category and Product Category Fact.
Does OBIEE support this kind of logical structure?
Case 2:
Tables:
- Product
- Product Fact 1
- Product Fact 2
Logical Join:
Product (Dim) -> Product Fact 1 (Fact)
Product (Dim) -> Product Fact 2 (Fact)
When I query columns from Product Fact 1 and Product Fact 2 (no column from Product), also fail. Any way to achieve so??
Thanks.

what error are you getting?

Model in BMM

Hi Everybody,
i have following model in physical Layer
I want to build that in BMM.
A -< B
A -< C
A -< D
E -< B
F -< B
B -< D
Actually, B is containing some facts. But if we drag the same physical layer to BMM, B is appearing as Dimension.
B, C, D are facts in other bi tool model.
Can you please tell me how to build the BMM Model
Thanks

Hi,
If B is a mix of dimensions and facts then your physical model is incorrect, it should be one or the other. Make an alias of B which will act as the fact in the physical layer, then it will be something like this:
A -< B_Fact
A -< C
A -< D
E -< B_Fact
F -< B_Fact
B -< D
B -< B_Fact (This will really be a one to one on the primary key for B)
When you drag this to the BMM layer it should work, as it will create a seperate logical table for the dimension and the fact version of B.
Regards,
Matt

Need Usage Pattern Report in OBIEE 11g

Hi All,
I have a requirement, where I need to know who logged in, to OBI Application in the last 3 months and who has not. How do I get that? Basing on START_DT in S_NQ_QCCT, I can put a filter with Start Date using BETWEEN RANGE, but since OBIEE logs in only those users who logs in, am not sure how to capture the records for non-logged in users during the same period. I have to report output in a pie chart with the percentages.
Any help would be highly appreciated, since its an urgent need. Thanks in advance.
Regards,
Vinag

You need to have all users in some custom table and customize the usage tracking BMM model by using outer joins so that you can see user who are not logged in.
If helps pls mark

Questions on OBIEE reporting against transactional Database

Hello guys
I have a situation where I have to bring in 5 tables from transactional database to do reporting on.. I looked at the tables, there are not typically dimension or fact tables. Each table contains information very specifically, at the same time all tables have numeric values that needed to be aggregated at the lowest level of that table..
I can join all these 5 tables together knowing the key columns. However, this is basically joining 5 tables with fact measures... Since this is transactional database which we don't own, so I can't create views nor changing the table structures..
I wonder how would I be able to configure such model in OBIEE?
Any thoughts?

Thank you for the reply..
The specific situation I have here is:
I have 4 tables that I am polling directly from the4 store's cash register (Oh yea, real real time), they are customer, receipt line, receipt header, inventory,
The physical join is: they are customer-----> receipt line<-----receipt header <------------inventory...
In the BMM layer, each of these 4 tables have columns that should be measures, such as customer.customer credit limit, receipt line. quantity, receipt header. reciept tendered, inventory.quantity on hand... They are more measure columns from each table, but I am just listing one from each..
In this case if i set the aggregation rule on each column, then the BMM model won't a star schema anymore because every table can be fact..
I thought about adding all the measure columns into 1 logical fact table, which then the logical fact will have 4 LTS: customer, Re Line, Re Header, Inv.. This will be fine if I am only dealing with 1 store... However, in reality there will be 50 stores with the same exact tables/structures but on different servesr. Therefore, the logical fact will end up with 4*50 = 200 LTS each has fragmentation defined "store number = NYC or something else"..
Just wondering if combining all the measures in one logical table is the only way to model this design or not? What would be the best thing to do if in this situation I have to report against multiple data source of the same tables in 1 subject area?
Please let me know your thoughts..
Thank you

Left Outer Join In OBIEE

Hi,
I have two Oracle tables in the physical layer: a fact table and a dimension table.
The fact table (FACT_TBL) like this:
CURR_MTH | PREV_MTH | CURR_GRADE | PREV_GRADE | CITY | VALUE
Jun | May | B | A | X | 5
Jun | May | A | A | Y | 10
Jun | May | C | C | X | 15
Jun | May | D | C | Y | 20
And the dimension table (GRADE_TBL) like this:
GRADE | GRADE_NAME
A | A
B | B
C | C
D | D
I joined those two tables above in the physical layer using GRADE field. Then, I created a business model using those two tables. I want to create a 4x4 matrix report with PREV_GRADE as the first column and the CURR_GRADE accross to the right like this:
CURR_MONTH PROMPT = Jun
PREV_MONTH PROMPT = May
CITY PROMPT = X
PREV_GRADE | A | B | C | D |
A | (null) | 5 | (null) | (null) |
B | (null) | (null) | (null) | (null) |
C | (null) | (null) | 15 | (null) |
D | (null) | (null) | (null) | (null)|
The problem is that I couldn't show all grade (A to D) in the PREV_GRADE column in that table. Every time I filter the CITY in the prompt to X, I always get only A and C. The B and D rows are suppressed. How do I achieve the left outer join that I wanted above? Thanks a lot for the help.

hi,
this can be done in BMM layer.
select the required BMM model. Right click and select "Buisness Model Diagram" and "whole diagram" .
now double click on the joining link and specify the type of join here. (assuming you have used "New Complex join" to join these tables).

Alias table

Hi All,
I have a situation where I have four tables having one to many database relationship.
There is no defined fact or dimension table.
X(1)-->A(N)
A(1)-->B(N)
B(1)-->C(N)
Can we create an alias for A and B to act as confirmed dimensions so we can pull all the
records from all the tables without any errors? I apologize for any confusion.
Thanks,
Virat

Hi,
No need for aliases, your describing one branch of a snowflake, simply flatten the snowflake (in your Physical model) to a star-schema in your BMM model (add the tables together in one or multiple Logical table sources).
Phy Layer :
X -> A -> B -> C
BMM Layer :
Logical Fact (X?) -> Logical Dim (A,B,C)
Regards
Alastair

Building Reports from Multiple Schemas

Hi,
I have a DB which contains 15 different schema. They contain the same tables and tables structure , with the data between different schemas being different. The user wants the same report to show data of each instance. I do not want to build 15 physical/bmm Models . Is there anyway this can be done? Please let me know the different options.
Edited by: user599926 on Feb 25, 2010 10:06 AM

Hi
You can build this as 15 Logical table sources and specify as Fragmentation source. Use a logical column to specify condition and combine sources which would work for you. Let us know result.
LG

How to create Analysis from two different subject areas with diff models

Hello,
I have problem with create analysis from two different areas with two different models (Model Star and Snowflakes) build in BI Admin tools.
First Model A = Subject Area A have one fact tables (with measures) and three dimensions tables (Model Star)
Second Model B = Subject Area B have one fact table (no measures) and five dimensions tables (Model Snowflakes).
I'm looking solutions how I can give good results. In this case I can't use UNION, INTERSECT etc., because this models haven't the same number column and data type?
whether it is possible to do in OBIEE?
Please help me
Kind regard
robix

Hi
One thing to understand is that if your are imagining having two Subject Areas (SAs) displayed and then dragging columns from each into your Request without any preparation, then no, that can't be done. You would need to do a join in the BMM layer and expose the column from the second Subject Area in the presentation layer.
But if you indeed interested in UNION (or intersect) queries, then do this:
Build your first query in the first SA in the usual manner. Note the number of columns and the data types of each column.
Now at the bottom of the Criteria workspace, there is a button called "Combine with similar request." Click it.
Select your second SA. You will note that the outlines of the columns in your first query appear. This is to remind you that the number of columns in this second query must be the same as in the first. It also shows the data type of each column for the same reason.
Once you select your columns and your filters, click on result. (Oh, you will also need to decide what kind of combination you want -- click on the dropdown.) The result will be a new Request that has data from both SAs in the query and combined in the manner you select (union, intersect, etc.)
That should give you a start.
http://obiee101.blogspot.in/2010/08/obiee-combine-with-similar-request.html -- Read this link this is good one to understand with screen shots. -- OBIEE 10g
http://oraclebiee11g.blogspot.in/2011/01/querying-across-multiple-subject-areas.html -- OBIEE 11g
Award Points it is useful.
Thanks
satya

How to build the best Purchasing Propensity Model

Hello, I'm trying to develop a Market Basket Analysis using over 90 different categories and I'm not quite sure which model would be the best option for this amount of variables?
Also I could use some direction on which algorithms and variables are the most relevant in the construction of a Purchasing Propensity Model (e.g. age, frequency of purchase, average ticket value, purchases in other categories, etc.)
Thanks in advance for your help

hi user,
Few things u should keep in mind
1) Take the tables and find the relation ship between those(pk-fk)
2) Identify Dim's and Fact's
3) After step2 definitely there will be a division of header level and detail tables
4) Dim1 having Business_Date and Fact1 having Business_date u can join in this way
5) Normally without ETL we may go for Pl/SQL or SQL's.Your Data model will come under this point
6)In BMM layer u can form Start Schema/Snow flake schema
Thanks,
Saichand.v

Error while moving objects from BMM to Presentation layer: 46036

Hi,
I created some aggregate logical columns in a logical fact table in BMM .
Next, I created a branch to already existing Dimensional hierarchy and used those aggregate columns in the branch hierarchy.
The Original Hierarchy was using columns from 2 Physical Dimensional tables.
While checked for global consistency , got the following error:
46036] Internal Assertion: Condition i + 1 < numKeys, file .\Src\SQLogicalRqtGenerator.cpp, line 645.
Where am I going wrong ?
Regards

Hi,
check if you presentation folder refers to the correct Business Model in the General Tab of the properties.
Regards,
Gianluca

Not able to find logical dimension tab in BMM layer in OBIEE 11G

Hi All,
Recently i ran in to a interested problem. We are using Oracle BI Administration Tool version 11.1.1.6.6, Repository Version:320.
I'm not able to see logical dimension tab to create dimensional hierarchy.
I followed normal method like all versions. Business model -> New Object -> (No logical dimension). I'm able to see logical display folder and logical Table.
I have only client tool on my machine. Please could anyone suggest me How should i resolve this problem.
Any help is appreciated.
Thanks,
Ashwini K.

Hi Srinu,
I joined all dimensions to facts, and i'm sure that i'm trying to create hierarchy for dimension tables. In oracle Administration Tool Repository version : 308, i'm able to see logical dimension tab at BMM-> New Object ->Logical dimension.
Any help

Difference between OBIEE Logical Modelling and physical modelling

HI.
what is the difference between logical modeling (BMM) and physical modelling(Physical layer) in obiee admin tool
please narrate it in definition format
thank you
rakesh

Hi rakesh,
Administrator tool consists of 3 layers
Physcial layer:-It is typically first layer built in the repository.Contains objects representing the physical data sources to which the BI server submits queries.It may conatin multiple data sources.The common way is to import metadat from different databases.You can define joins after import if they are not defined.
Business model and mapping layer:-It is defined as business or logical models of data and specifies the mapping between the physical and business layers.This is where the physical schemas are simplified to form the basis for the end user.We can say data is seperated into facts and dimension tables
Presentation layer:-simplifies and organizes the business model to make it easy for users to understand and query
go through the user guides for all the definition and subject
Hope it helps you.my points?
By,
Kranthi
Edited by: Kranthi.K on Sep 1, 2009 10:22 PM

Bmm Model

Similar Messages

Maybe you are looking for