Question on Logical table Sources

Good day!
I would like to ask for your advice on an bottleneck I am experiencing in a repository implementation in Siebel/OBIEE.
I am to create a logical table which sources comes from two different tables(Physical table A and Physical table B).
I am trying to union the contents from table A and table B so I planned to use fragmentation content.
My problem is that there is no way to distinguish the difference between the records coming from physical table A and table B.
I created two logical tables sources and specified the same condition for both but the server would automatically choose the first one and would not include the records from the second logical table source.
Is there any other way I could combine the records from table A and table B in Business Model and Mapping layer?
Please, I would like to know your thoughts.
Thanks,
Ivan

Hi Ivan,
I dont think there is other way which is equable to fragmentation content functionality.
I would suggest to try with any column is not null or most satisfied condition so that both contents present in the query.
If helps pls mark

Similar Messages

Question related to Logical table source

Hi All,
I have very basic questions -
(1) When do we create multiple logical table sources within a dimension?
(2) If a dimension has more than one logical table sources , when do we need to map the underlying table to other underlying table of logical table sources.
For exp consider these tables XLE_Entity_Profile, XLE_REgistration , AP_Invoice_ALL
AP_Invoice_All is a fact table. relationship between XLE_Entity_Profile & XLE_REgistration is (1:M).
Join info is as below -
(a) XLE_Entity_Profile.Legal_entity_id = AP_Invoice_ALL.Legal_entity_id
to get the registration of LE the where clause is as below-
(b) XLE_Entity_Profile..Legal_entity_id = XLE_REgistration.Source_id and XLE_REgistration.Source_table = 'XLE_Entity_Profile'
I have created the alias of XLE_REgistration as XLE_REgistration_LE.
With in a dimension , i have 2 logical table source - XLE_Entity_Profile & XLE_REgistration_LE.
logical table source 'XLE_REgistration_LE' has where clause 'XLE_REgistration.Source_table = 'XLE_Entity_Profile'
When i query , LE name , LE Registration Name, i get error like -
State: HY000. Code: 10058. [NQODBC] [SQL_STATE: HY000] [nQSError: 10058] A general error has occurred. [nQSError: 14070] Cannot find logical table source coverage for logical columns: [LE_NAME]. Please check more detailed level keys are mapped correctly. (HY000)
Le Name is from XLE_Entity_Profile and LE Registration Name is from XLE_REgistration_LE.
But when using the properties of logical table source ' XLE_REgistration_LE' , i map it to 'XLE_Entity_Profile' ,i get the correct result.
I am not able to understand why am i getting error in first way of modeling.
Thanks , Ashish

Hi Ashish,
first about logical table sources (lts): you can create different lts for aggregation and/or fragmentation. Aggregation means that data is aggregated on another level among different physical tables. Fragmentation means that the content is different over different tables (other rows).
In your case, I think the problem is that your dimension is not denormalized, which results in a snowflake.
What I understand is that you have the following (physical diagram):
Invoice (fact table) >----- Entity (Dim) >----- Registration (Dim)
You have the following joins:
invoice.entity_id = entity.entity_id
and entity.entity_id = registration.source_id and registration.source = 'something'
First idea:
I would create the following join (in physical diagram):
Invoice (fact table) >---- Registration (Dim)
Where:
invoice.entity_id = registration.source_id and registration.source = 'something'.
Then in your Entity dimension you should create a hierarchy:
Grand Total Level
Entity
Registration.
In your dimension you should create the first lts: Entity
Set the aggregation content of this lts to Entity.
This lts contains only one physical table.
Map only the entity columns to the Entity physical table.
Then create the second lts:
Entity and Registration.
Set the aggregation content of this lts to Registration.
This lts must contain two physical tables, Entity and Registration.
Map the entity columns to the Entity physical table and the registration columns to the registration physical table.
Let me know if it works or not.
Regards,
Stijn

Problem: 1 physical table -- multiple logical table sources

Hi,
I'm quite new to BIEE and setting up my repository.
So I have a question, if the following scenario is possible:
Physical Layer: TABLE_A: COL_A, COL_B, COL_C
TABLE_B: COL_D, COL_E, COL_F
Join TABLE_A.COL_A = TABLE_B.COL_D
In Business Model I have a Dimension Table with TABLE_A as datasource with fields DIM1 (COL_B).
The Fact Table (MEASURE) would have twice TABLE_B as data source with different where-clauses on COL_F and logical table columns (ATT1 and ATT2) of value COL_E.
So far I have created everything and the consistency check shows no errors or warnings, but I get an error in Answer: Incorrectly defined logical table source (for fact table MEASURE) does not contain mapping for [MEASURE.ATT1, MEASURE.ATT2], when I creating an report showing DIM1, ATT1, ATT2.
Isn't it possible to have one physical column used as multiple data source?
I know it 's working, when I create the physical table twice ... but maybe there's a solution for business model.
Thanks
chrissy

Hi mengesh,
that's what I also tried, but it's always returning me the same error.
I know it would work, when I import the physical table twice or more, but that's not what I want to do, because at the end I have 10 or more fields based on this one physical table. There's one field indicating what value is contained in the record, this means:
COL_F | COL_E
1 | customer name
2 | customer number
3 | customer branche
4 | salesman
5 | date
6 | report number
etc.
I don't think it's usefull to import the physical table as often as I need this field. So I want to divide it in business model.
thanks
chrissy

Multiple Logical Table Sources In BMM Under 1 Logical Table In OBIEE 11g

I have a question and even after doing a lot of search on google i could not find any article so through to ask you :-
I want to create a Logical Dimension table
This logical dimension has columns coming from 2 different physical tables, TableA and TableB. The relationship between TableA to TableB is 1 to zero.
There are 2 ways to create the logical dimension :-
1) Go to the logical table source properties of TableA-->then there + sign click and then add TableB right outer join with tableA.
Here in the logical table source we see only one table tableA.
2) Drag the tableB column so that we have 2 logical table source.
Please try this to explain without using the concept of fragmentation and without considering the fact tables. I have this question for simple logical dimension tables only.
I wanted to know which is the right way and what requirement/factor decides what to do ?

Check this post
Business Model - Logical Table Source
Let me know for Qs.
If helps mark as correct :)

How to "think" use of multiple logical table sources

Hi everybody,
i'm new to OBIEE and its paradigm. I'm trying to build a prototype with multiple physical sources: ERD and a Data Mart; I need OBIEE to go to the Data Mart model at the need of aggregated info and to go to the relational model at the need of transactional info. I need to do that working with dimensions, assigning different sources to each logical level of the dimension.
My problem begins when thinking how to actually do that. Here is my question:
- Should I add the relational source in the data mart logical table source folders, mapping the same information with joins? If so, how does dimensions work from that point? (drill down)
thanks in advance!

I'm a little confused
Is the problem
1 - You want to Access a table in an existing 9i database from a different 10g database
or
2 - You have a new 10g database as a copy of a 9i one, and a table called system has gone missing?

Repos - logical table source does not contain mapping

Hi,
I have a repository question?
I do not have a fact table in my physical layer. Do I have to have one? I thought the fact table got created in the BM.
I do have a date table in the physical layer, but it isn't joined to any other table.
I added the PK from the date_dimention to my logical fact table.
when I open the properities of my logical date table it is mapped to the physical date table.
Now in answers, I can query the date_dimension without error.
But, when I query the fact table using the date key, I get this error.
Error Details
Error Codes: OPR4ONWY:U9IM8TAC:OI2DL65P
State: HY000. Code: 10058. [NQODBC] [SQL_STATE: HY000] [nQSError: 10058] A general error has occurred. [nQSError: 15018] Incorrectly defined logical table source (for fact table CRTK_FACTS) does not contain mapping for [CRTK_FACTS.MASTER_AI_ID, CRTK_FACTS.INT_DOC_ID, CRTK_FACTS.DATE_KEY]. (HY000)
SQL Issued: SELECT CRTK_FACTS.MASTER_AI_ID saw_0, CRTK_FACTS.INT_DOC_ID saw_1, CRTK_FACTS.DATE_KEY saw_2 FROM CRTK ORDER BY saw_0, saw_1, saw_2
Anyone have any suggestions on what I am missing?
thanks,
Kathy

I don't have a fact table in the physical layer. Do you mean join between the date table and other dimension tables?
or do I have to build the fact table in the physical layer?
I have not built any hierarchies yet, working on it.

Logical table source

Hello!
Can someone please explain difference between dragging and dropping a physical column from a table in Physical Layer to a Logical Table in BMM layer and ragging and dropping a physical column from a table in Physical Layer to a Logical Table Source? When to use these two options?
Thanks,
-D

Hi,
First you have to read the OBIEE Administrator concept and will clear you doubt,
OBIEE Administrator, Windows based client tool to develop and manage Metadata repository. Repository changes can be done in either online or offline mode.
Online mode is when BI Server is actively reading the repository file (when a user is logged in).
Offline is when BI Server is idle. Best practice is to develop in Offline mode.
Minor changes can be made when in online mode. Administrator uses ODBC connectivity to connect to physical data sources, uses Native Drivers for Oracle and DB2.
OBIEE defines three layers of abstraction.
Physical Layer – connections to physical data sources (OLTP, OLAP, Flat Files, Oracle, SQL*Server, Teradata, etc.).
Business Model Layer – Brings together data elements from Physical sources into a logical Star-Schema representation. This is where business rules and definitions are setup.
Presentation Layer – An abstraction from the Business Model, defines exactly how users will see tables and fields.
Connection Pool is a connection to a Physical data source. Multiple connection pools can be created to the same source to improve end-user query performance.
All tables in the Physical layer must have a Primary Key column. Column properties can be edited to set the Primary Key. Physical joins need to be established between tables, using Foreign Keys and Primary Keys. Right-click on Connection pool to create the Physical diagram joins.
A Business Model can be created by drag and drop of objects from the Physical Layer. It’s a logical Star-Schema representation of the Physical objects. Tables from different sources can be merged to form one Logical table. These sources are called Logical Table Sources.
Logical Fact tables hold Metrics; every Metric must have an aggregation rule. Hierarchies can be setup on dimensions for drilldown functionality.
It is important to join logical Facts and Dimensions together, much like the Physical Diagram. Right-click on Model, click Business Model Diagram to create joins. Use Complex joins in the Business Model.
Presentation Layer is exactly what End-Users will see in BI Answers, contains multiple Subject Areas that answer specific Business questions.
Thanks,
Balaa...

Business Model - Logical Table Source

Can someone give me clarification on exactly when you would follow scenario 1 when modelling a logical table and exactly when you would follow scenario 2, and what the key difference in behaviour between the two is. It would help if someone could illustrate with sql table joins equivalent.
Scenario 1
You drag a second physical table's column onto the logical table. This causes an additional field to appear in that logical tables list of columns sourced from a different table to the original and causes a second table to appear under the original table in the source folder.
Scenario 2
You drag a second physical table onto the existing logical table source of a logical table. The physical table source appears on the surface the same as before, but when you examine its properties you see that the second table has been joined to it.
Thanks for your input,
Robert.

Hi Robert,
This is quite an interesting question, I would say :). As per me, here is why
As you know already, BI Server generates intelligent queries based on our logical mappings in the BMM layer and there is only one primary goal for our BI Server Engine, generating an optimistic physical query.
Scenario 1 : Two logical sources created.
Two logical sources mean, there are two different ways the logical table(or same logical column) could be arrived at.
Ex 1:
You have a model, where you have Sales numbers and 4 levels of product hierarchy (lowest P4 -> P3-> P2-> P1) with sales at productlevel4.
And, You have created aggregated tables for sales numbers at level2 of product dimension also.
So, in this case the same product dimension would have two logical table sources, just because Level 1 Product can be derived from Aggregated product dimension source at level2 or detailed product dimension source at level4. And, it is the BI Server which would determine from where it has to pick (Of course there are some other criteria to control it).
Ex2:
Here even sales numbers are available at different granularities viz product level2, product level4. So for sales logical fact, we end up creating two sources just because there are two completely different sources with different granularity.
Scenario 2 : Two columns created, but only one logical source with a join in it.
Say, in the above example you have a snowflake product manager dimension to the product dimension. So, when you pull Product Manager into your product dimension (collapsing snowflakes), you end up creating just one logical table source with an inner join of ProductManager dimension and product dimension because product manager could be arrived only by going through products.
So, if you have report on sales by product manager, you would see query with an inner join of sales fact to product and inner join(this is based on the join type you set in this logical source) of product to product manager.
Hope I was clear enough and this helps.
Thank you,
Dhar

How would multi Logical Table Source work?

Hi All,
I am facing a problem about multi Logical Table Source. Can I build a Business Model with a Logical Fact Table that have two Sources? Taking the below description as an example:
Table Name
Columns
Description
Fact
TIMECD(2013Q101, which join to Time.TIMECE), DATA
Fact_Aggr
TIMECD(2013Q1, which join to Time.PERQUARTER), DATA
Aggregate table for Fact and sum Monthly data to Quarterly
Time
TIMECD, YEAR, PERYEAR, QUARTER, PERQUARTER, MONTH, PERMONTH
I tried to create a BMM that contains two Fact Table in the same Logical Fact Table. But BIEE shows the DATA from Fact even if I drag PERQUARTER into the report.
The following is my question:
1. Can I have a Logical Fact Table that contains Fact and Fact_Aggr?
2. How will BIEE determine which DATA should be used?
Best Regards,
Martin

The basic rule for one LTS vs two is: If you use table A AND Table B together, then one LTS. If you use table A OR Table B, then 2 LTSs.
So when you have an aggregate fact table, you use either the Agg or the base fact, meaning you should have 2 LTSs. OBI understands how to use one or the other through a variety of rules, but the main piece of config you need to have is proper levels defined on each and every LTS's Content tab. In your example, you will need to make sure the grains of your aggregate and base fact table match the time dimension properly; thus normally you have a day table and a month table in the time dimension, each joining to the appropriate Fact LTS. If you join a month level aggregate to a day time table your results will be incorrect. OBI knows that month is higher in the hierarchy than day, so it will prefer to use the set of Month LTSs when it can.
Jeff M.

Logical Table source source query

In OBIEE 10g we can have multiple logical table sources and we can also add multiple tables into a single logical table source(logical table source source). I wanted to know the difference between doing so and having multiple logical table sources for each logical source.
Hope I made myself clear.
Cheers
Rem

Hi Rem,
When data is duplicated across different physical tables add them as separate LTS with column mapping pointing to most economical sources. Specifying the most economical source is about the idea that a single column exists in more than one table, based on the column mappings BI server picks up those LTS's which could satisfy the request with minimal joins.
When the data is not duplicated add them in a single LTS source. When the physical sources are added in a single LTS, you have the flexibility of using outer joins. But specifying a join as outer join makes BI Server to include this source even if its not required otherwise when the join is inner, the sources will not be included if not required to satisfy the query.
Hope this helps.
Thanks!

Filter on the same logical table source on two logical tables is not working properly

Hi,
In the RPD, assume that I have a physical table named Employee_fact, created an Alias called D0 Employee Fact
In the BMM I've two logical tables called Employee_fact_Type1 and Employee_fact_Type2
Each of these logical tables has the same Logical table source - D0 Employee
But in each logical table source, I've a filter in where clause
such as, for Employee_fact_Type1 employee_type = 'XXX'
for Employee_fact_Type2 employee_type='YYY'
These two facts are present in the same BMM and each of them are connected to two dimensions
such as
Dept_type1 <- Employee_fact_type_1 -> Tenure_type1
Dept_type2 <- Employee_fact_type_2 -> Tenure_type2
In analysis, When I query the columns from Dept_type1 and Tenure_type1, the database query generated shows the condition employee_type='YYY' instead of employee_type='XXX' .
But, adding the columns from Dept_Type1, Employee_fact_type_1, Tenure_type1 it shows proper filter.
I checked the relationships, presentation layer for the sources of presentation tables and everything looks good.
Can anybody tell me what goes wrong here?

did you set the content level as well?

Dynamically change the Priority Group of Logical Table Sources in OBIEE 11g

Hi All,
I have 2 Logical Table Sources(LTS 1 and LTS 2 for a Logical Table in BMM Layer).
Example: Logical Table : Sample
LTS Source 1 : Sample 1(Priority Group Set to 1)
LTS Source 2 : Sample 2(Priority Group Set to 0)
I have set the Priority Group of Sample 1 LTS Source to 1 and Priority Group of Sample 2 LTS Source to 0.
I need to dynamically change the Priority Group of Sample 1 LTS Source to 0 if my role is DEVELOPER where role is a column in database.
If my role is not equal to DEVELOPER then the Priority Group of Sample 1 LTS Source will remain same(1).
Please suggest how can i achieve this.
Thanks,
Soukath Ali

hello Soukath Ali,
didi you find a way to dinamically changing Priority Group?
thanks,
Maria Teresa Marchetti

Max number of logical table sources

Hi,
I have one logical table based (a fact table) on 8 logical table sources. This is done to simulate some partitioning as my customer does not have an partition set up on data base level.
Anyway my challenge is that a request fired in answers does not seem to take all logical table sources into account when building the physical SQL.
The missing logical table source (the one not being part of the physical SQL) does not differ from the other logical table sources.
Is there a limit of how many logical table sources BIEE can handle under one logical table?
BIEE version: 10.1.3.4

Alastair,
thanks for replying.
The logical tables sources partition the data by year. The aggregation level is the same for all LTS, the check for using this with other sources at this level is set and I've tried to force a result from every LTS (not combined but each LTS for itself). Forcing the results is done by using a dashboard prompt in answers filtering on year.
The whole 'LTS-setup' works in combination with Oracle Warehouse Builder. What I mean by this is the following: In OWB I split the data from one table into 8 different tables, each containing 3 years of data (except for the oldest data where the number of years and amount of data will grow over time).
This is done by comparing values of a year column with the actual year. E.g. the most actual table would contain data for the years 2010 - 2008, the next would contain data for 2007 - 2005, and so on.
Now as the years in the different tables will change, I decided to establish an equaly changing counterpart in BIEE. Basically I defined some repository variables: actaul_year, three_years_ago, six_years_ago, 9_years_ago, etc.
The fragmentation content of the different LTS is defined like this:
most actual data: MyModel.Year.Year >= VALUEOF("3_years_ago")d
next table: VALUEOF("3_years_ago") > MyModel.Year.Year AND MyModel.Year.Year >= VALUEOF("6_years_ago")
next table: VALUEOF("6_years_ago") > MyModel.Year.Year AND MyModel.Year.Year >= VALUEOF("9_years_ago")
etc.
The funny thing is that the missing table is the one containing years 13 - 15. All the other tables (e.g. 16 - 18 years and 19 - 21 years) are represented in the physical SQL. This particular table is not even represented in the physical SQL.
I haven't yet tried to take one of the 'working' tables out to see if this helps (getting an idea if there is a limit on number of LTS per logical table).
Any idea?
Thanks
regards
Andy

Multiple Logical Table Sources vs Single Logical Table Source

OBIEE 11g. I am totally confused. Can someone help me with the following for me.
I have seen logical table sources being used in 2 ways. I have a 2 sources tables: CUSTOMER & ADDRESS. In physical layer these 2 tables are linked by CUSTOMER _ID in a 1:M relationship. i.e. A customer can have many addresses.
Scenario # 1: In BMM the two tables form a logical table called CUSTOMER with 2 different logical table sources which are CUSTOMER & ADDRESS.
Scenario # 2: In BMM the two tables form a logical table called CUSTOMER with 1 logical table source called CUSTOMER only.
What is the difference between the above 2 scenarios and which one is better to use when creating the logical table source.
Regards.

Scenario # 1: In BMM the two tables form a logical table called CUSTOMER with 2 different logical table sources which are CUSTOMER & ADDRESS.
-->In this case; based on the columns select source is defined in the physical query
possible case would be CUSTOMER , ADDRESS, and CUSTOMER & ADDRESS
-->Used in DE-normalized scenarios
--> BI uses intelligence based on content tab settings
Scenario # 2: In BMM the two tables form a logical table called CUSTOMER with 1 logical table source called CUSTOMER only.
-->In this case; CUSTOMER properties->General tab used both CUSTOMER & ADDRESS, in your physical query both tables present irrespective of column selection
-->Used in normalized scenarios
-->Forcing BI to follow our way, since we wont set content tab settings
Hope this helps

Time series functions are not working for fragmented logical table sources?

If i remove that fragmented logical table sources, then its working fine.
if any body know the reason, please let me know.
thanks and regards,
krishna

Hi,
is because the time series function are not supported for the framentation content, see the content of the oracle support:
The error occurs due to the fact the fragmented data sources are used on some Time series measures. Time series measures (i.e. AGO) are not supported on fragmented data sources.
Confirmation is documented in the following guide - Creating and Administering the Business Model and Mapping Layer in an Oracle BI Repository > Process of Creating and Administering Dimensions
Ago or ToDate functionality is not supported on fragmented logical table sources. For more information, refer to “About Time Series Conversion Functions” on page 197.
Regards,
Gianluca

Question on Logical table Sources

Similar Messages

Maybe you are looking for