One dimension used twice in a cube

I have one dimension, but it has two meanings. How can i use the same dimension twice in the same cube as it was two separate dimension?
Example:
There is dimension "Region". The fact table has two keys. First key means - "region of birth", and second "region of living". So one dimension has two meanings.
I tried put the same dimension twice with two different aliases in one cube. OEM shows such cube as valid, but querrying it through OLAP API, or building analytic workspace based on this cube is impossible (error ORA-1007).
The only workaround i see is to build two separate but identical dimensions.The bad side of this is that i have to create two separate but identical tables of data for those dimensions (i can't create two levels from the same column).
Raf
i work with Oracle9iR2

i did the same on Oracle9iR2 with OEM 9.2.0.1.0.
Creating the cube with two aliases for one dimension was ok.
But creating MOLAP cube from that source cube using AWM was impossible .
Also querrying that cube (the one created in OEM) for example through Discoverer or Excel is imposible.
I searched forums on that question, and some people had the same problem, but no answer.
Raf

Similar Messages

How to delete the members in one dimension use the maxl script

i have question that i want to delete the members in one dimension useing the maxl script, but i do not know how to do it. can the maxl delete the members in one dimension? if can, please provide an sample script, thank you so mcuh.

MaxL does not have commands to alter an outline directly, except the reset command which can delete all dimensions but not members selectively. The best you could do would be to run a rules file (import dimensions) using a file that contains the members you want to keepload rule for the dimension. As typical the warning is to test this first before you do it on a production database

Moving the object from one dimension to other in the cube in BI7.0

Hi Gurus,
I am new to BI7.0
We have a requirement wherein we need to move the object from one dimension to other dimension within same Infocube.I tried doing 'cut' and 'insert' from the context menu of the dimensions, but it didnt work.
Please let me know how do we do that..
Thanks,
Shalaka.

Hi,
Use Remodelling opttion avialable .For more info, earch the forum and go through SAP standard help.
With rgds,
Anil Kumar Sharma .P

Referencing two fact keys from one dimension using Dim.Key = Fact.Key1 OR Fact.Key2 possible ?

Hi all
I need your help for an approach to the following problem. Lets say that i have the following dataset
AccountMonth
ActivityMonth
Amount
201307
201301
2.500
201307
201304
600
201307
201305
900
201307
201306
4.000
201307
201307
500.000
201308
201305
500
201308
201307
400
201308
201306
300
201308
201308
400.000
201309
201307
300
201309
201302
500
201309
201309
100.000
201310
201305
400
201310
201309
50.000
201310
201310
200.000
In my cube i want the user to select the Accountmonth from one time dimension, but i want some logic to also apply for the ActivityMonth. For one measure if I for example select 201307 i'll get this amount
AccountMonth
ActivityMonth
Amount
201307
201301
2.500
201307
201304
600
201307
201305
900
201307
201306
4.000
201307
201307
500.000
This is easy - Accountmonth dimension is related to AccountMonth column in the fact table. Now i also need another measure, where the month selected needs to be meet for both accounting and activity month. This is also easy, as i can just reference both
columns in the fact table. Result in this case is then only one row:
AccountMonth
ActivityMonth
Amount
201307
201307
500.000
Now the tricky part comes when I start to select more months, because the amount i need is where month combinations are shared, so if i for example choose 201307, 201308 and 201309 i don't just want
AccountMonth
ActivityMonth
Amount
201307
201307
500.000
201308
201308
400.000
201309
201309
100.000
I actually want the amount from which the months is in either accounting or activity, like this
AccountMonth
ActivityMonth
Amount
201307
201307
500.000
201308
201307
400
201308
201308
400.000
201309
201307
300
201309
201309
100.000
I know how to solve it either by having two time dimensions or using MDX, but user should only have to select from one time dimension. So far, solving it using Non-empty/existing in MDX is not performing fast enough. Feedback will be appreaciated if you
have an idea of how to solve this.

So far, solving it using Non-empty/existing in MDX is not performing fast enough. Feedback will be appreaciated if you have an idea of how to solve this.
here are some links about performance tuning
http://www.mssqltips.com/sqlservertip/2565/ssas--best-practices-and-performance-optimization--part-1-of-4/
http://technet.microsoft.com/en-us/library/cc966527.aspx

Department Dimension used twice for Contribution Logical Source

How do you join the same hierarchy twice to a Logical Table given the following logical columns?
Source Department, Destination Department, Amount Contributed
Sample Data
Corporate, Sales, 1,000
Corporate, Development, 2,000
Hierarchy
Top
|- Department
Corporate
|- Sale
|- Development
|- Support

Hi,
What you need to do is to add the (physical) location dimension table to the logical table source of the demographic dimension, for example by dragging it from physical layer on top of logical table source of demographic logical dimension table in bmm layer
Regards,
Stijn

Creating a cube with more than one dimension

I have been able to create a cube with one dimension using our own data. I am able to view data from this cube in Cube Viewer and in a presentation created with BI Beans in JDeveloper.
However, I have been unsuccessful in doing this when creating a cube with two dimensions. What am I missing?
I have been using OEM to create the dimensions and cubes, etc. in a 9.2.0.4 database.

You can use Analytic Workspace Manager 10.1.0.4 to create analytic workspaces (MOLAP) in Oracle OLAP 10.1.0.4. The Model View in this GUI tool utilizes the OLAP AW Java API that was introduced in Oracle OLAP 10g. This Java API fully abstracts the logical dimensional model from physical design.
OWB Paris, which is currently in beta, likewise uses the OLAP AW API to create AWs. In addition, it can create ROLAP cubes via the OLAP Catalog CWM2 APIs. This is a change in APIs as the previous version used CWM1/Lite like Enterprise Manager. Also, for ROLAP cubes OWB Paris will automatically include MR_REFRESH in its scripts and will call the appropriate DBMS_ODM package in order to create materialized views.
As for tutorials for creating a star schema, check with the Oracle Warehouse Builder forum.
OWB Forum:
Warehouse Builder

Can we have more than one dimension in a Report ?

Dear Friends,
I am trying to build a group matrix report from more than one dimension using Express Data.
Is this possible ? If yes can you please guide me as to how to do this.
Thanks in advance.
Regds,
Ramakrishnan L

Please refer to the section" Building a Report Using an Express Data Source example" at
http://otn.oracle.com/products/reports/htdocs/getstart/examples/index.html
Thanks,
Siva

Use one dimension to feed 2 other dimensions

Is there a way to use one dimension to push data into two other dimensions using push logic? The client has one dimension in App1 that is a combination of 2 dimensions in App2 ie the Costcenter dimension in App1 is a combination of Plant and Account in App2.
Now they want to push data from App1 into App2 using the Costcenter dimension to populate the Plant and Account.
I have Plant and Acct properties in the Costcenter dimension for mapping.
I tried to use *Rename_dim CC= CC.Plant and *Rename_dim CC=CC.Acct but would get errors when validating the logic script.
Any help would be greatly appreciated.

Parameters
COSTCENTER: 70101014
When I test the logic using the logic debugger, I get the error:
Validate member failed:
70101014
70101014
70101014
70101014 on COSTCENTER dimension
When I run the logic thru data manager, the package is successful but the end of the detail status log is 'No records to process'
The push logic should ideally take the Costcenter 70101014 from App1 and take the first 4 char (7010) and record to the Plant dimension in App2 then take the last 4 char (1014) and record to the Account dimension in App2
Logic script:
*DESTINATION_APP= APP2
*RENAME_DIM CC = ACCOUNT
*RENAME_DIM CC= PLANT
*WHEN CATEGORY
*IS "AOP"
*REC(CC=CC.RPT_A,CC=CC.PLANT,CATEGORY=AOP)
*ENDWHEN
*COMMIT

Using different logical tables in one dimension

Hello.
I have two fact tables and two dimension tables Country and Region. One fact table depend on one dimension table like it shown in this link - http://file.qip.ru/file/121151603/3b4ebdba/2_online.html. Table Country is received from table Region by DISTINCT.
At the presentation I have to realize one column in Presentation Catalog for both columns Country Name (or Country Id) from Country and Region, the user won't know that one fact table depends on COUNTRY and other on REGION.
Also I have to realize the dimention "Geography" (Country->District->Region) for fact table SALE_REGION.
Could Oracle BI automatically choose from which table get data and if it can, how I should make the logical model?
Thanks.
PS.
The best variant I realized it this - http://file.qip.ru/file/121154314/8d242b81/3_online.html where District is the shared child for second Country level, but it doesn't work properly.

Y don't u use union request in first request details of fact1 and its related dimensions and in 2nd request fact 2 and its related dimensions.
I guess this should work
Thanks
Subhash

Can one ODI procedure be used twice with different option in one package?

I have a procedure with three check box type options. This procedure is used twice in a package. I tried to set options differently in the package, but once I set up the options for the second call, the first one will be updated to the same as the second one automatically. Is the option supposed be used as parameter? If it is true, why I cannot call procedure twice by using two different parameter settings?
Thanks

Create two scenarios of the same procedure. While generating scenarios change the options as necessary. Then in the package call the scenarios instead of the procedure. That should work.

SSAS : one dimension instead of 3 times the same

I want to make a cube that show, for each factory, how many units produced and how many units transfered between 2 factories by month.
I have a DimFactory dimension that represents factories.
My 1st fact table (FactProduction) is what have been produced in a factory. So I have a FactoryKey that links my FactProduction to DimFactory.
My 2nd fact table (FactTransfer) is what have been transfered (i.e. shipped) between 2 factories. I have 2 foreign keys: FromFactoryKey and ToFactoryKey.
Both are linking my FactTransfer to DimFactory (so for each record I have a FROM and a TO).
I created a cube that uses both fact tables but of course when I do so, I have to use 3 dimensions (each of them being DimFactory):
Location      (FactProduction.FactoryKey -> DimFactory.FactoryKey)
LocationFrom (FactTransfer.FromFactoryKey -> DimFactory.FactoryKey)
LocationTo    (FactTransfer.ToFactoryKey -> DimFactory.FactoryKey)
I would like to use only one dimension (Location) and using Calculated Fields I would have 3 measures:
Quantity Produced
Quantity Transfered From
Quantity Transfered To
So for the currentmember in DimFactory I would have 3 columns.
MONTH   |   FACTORY       |   Produced   |   TRF FROM   |   TRF TO
===============================================================================================
JANUARY       NEW YORK       45,600           23,450           12,344
FEBRUARY   MIAMI           20,200           12,233           5,673
What kind of calculation I must do to get these measures (and be able to use one DimFactory instead of three)
Thank you for your help

I found a way to do what I was looking for, by splitting my records in half.
So my fact table now has 2 records for each transfer, with only one FactoryKey with a column telling what kind of record it is (TO or FROM).
That way I can have a calculated field saying "if it is a TO, then TRANSFER IN QTY is QTY else TRANSFER IN QTY is 0 and same logic for a FROM"
In my cube I have only one dimension (DimFactory) and for each factory I have 2 columns : TRANSFER IN, TRANSFER OUT.

BI beans does not use QUERY RERWITE for cube

Hello!
First of all I would like to say big thanks to Keith for help on dimension rollup. It works now.
I am creating a pilot environment with
Oracle 10.2.0.3
OWB 10.2.0.3
SS Add-in 10.1.2.2
and one cube ST_R and three dims
DIM_R_NOMA for products
DIM_R_CIDI for stores
DIM_R_TIME for time
Now I have deployed dimensions and cube to CWM2. Dimensions are working quite well. I would say even better than in MOLAP. About 1 second for rollup on 2000 members in level.
Now I am facing second problem. I can not force BI beans (which is used in SS Add-in and Disco and OWB browser) to use summaries prepared by DBMS_ODM package.
1/ I have prepared MV LOGS for dimension and fact tables
2/ I have prepared MV for dimensions using DBMS_ODM
3/ I've prepared materialized view for cube aggs. with
DBMS_ODM.CREATESTDFACTMV('WWHH','ST_R','ST_R.sql','C:\TEMP',true,'FULL');
CWM2_OLAP_CUBE.set_mv_summary_code('WWHH','ST_R','GROUPINGSET');
as described in OLAP REFERENCE on DBMS_ODM page.
I explained all my MVs - they seem to be OK. They support
REWRITE_GENERAL,
REWRITE_FULL_TEXT,
REWRITE_PART_TEXT
There are no support for PCT rewrite etc.
My user 'WWHH' has a privileges:
ANALYSE_ANY
QWERY_REWRITE
GLOBAL_QWERY_REWRITE
My database has setting
QUERY_REWRITE_ENABLED=true
Stale_tolerated=enforced
all MVs and tables are analyzed.
I do not use parallel settings on tables,MVs.
to do some further analisys I've enabled olapcontinuous_trace_file. It generates some useful SQL statements from BI BEANS in UDUMP. These statements DO NOT resolve in MAT VIEW in explain plan
Questions:
Are there any settings for BI BEANS to turn on/off?
Are there any other packages to create MVs?
How to explain WHY MVs are not used?
THanks everybody for cooperation.
Regards,
Kirill Boyko

Keith,
Thank you for response.
I refreshed metadata, but no result. Again, dimensions are OK, but not cube. I am attaching explanation why it did no rewrite. This explanation is coming from
dbms_mview.Explain_Rewrite
QSM-01150: query did not rewrite
QSM-01263: query rewrite not possible when query references a dictionary table or view
QSM-01284: materialized view MV1125 has an anchor table DIM_R_TIME_V not found in query
QSM-01102: materialized view, MV1125, requires join back to table, DIM_R_CIDI_V, on column, REGION_ID
QSM-01219: no suitable materialized view found to rewrite this query
QSM-01284: - is wrong. I have DIM_R_TIME_V in MView.
Here is my MVIEW script from database:
SELECT
FROM
WWHH.CAWHTB2 a,
WWHH.DIM_R_CIDI_V b,
WWHH.DIM_R_NOMA_V c,
WWHH.DIM_R_TIME_V d
WHERE
b.STORE_ID = a.CIIDCA AND
c.ARTICLE_ID = a.NMIDCA AND
d.DAY_ID = a.CLIDCA
GROUP BY GROUPING SETS ( ...)
Could you tell me where to check in metadata that MVIEW is correctly setup?

Multiple hierarchy in one dimension

hi ,
i have a dimension with 4 levels and 2 hierarchy . one hierarchy uses all 4 levels (default hierarchy) and one uses 3 levels .
i have 2 cubes , one cube needs to use hierarchy 1 from that dimension and there is no problem on validation time .
one cube needs to use hierarchy 2 of that dimension when i uses 'deploy all ' as deployment option , OWB says to me that i have to use last level of that dimension . I use relational implementation .
can i use diffrent hierarchy of 1 dimension in diffrent cubs ?
thank u in advance for your advise
Padideh

Hi Sankar,
You can define Employee, Organization, Physical, Customer & Product as navigational attributes of Role and in the query enable the hierarchy for Role. Then pull Employee, Organization, Physical, Customer & Product under Role this way you will be able to display, the hierarchies of Employee, Organization, Physical, Customer & Product under the nodes of Role hierarchy in the order in which you place them in the report.
Hope this helps.
Regards
Kumar

Reference same dimension multiple times in a cube

Hi,
I'm creating a cube with Analytic Workspace Manager. The cube needs to reference same dimension more than once. For example, in a movie ticket booking cube, there's a booking date and a show date , both should reference same pre-created TIME dimension. Because one dimension can be added to a cube only once, now my solution is to define 2 dimensions, i.e. BOOKING_TIME and SHOW_TIME, both are based on same underlying table and have exactly same levels/hierarchies/mapping, and then add them to the cube. I believe there must be smarter approaches out there, please advise.
Thanks!
Jim
Edited by: user11955172 on 2012-12-10 上午2:36

For hand written/custom sql going against the Cube:
You're right that you can use one of the 3 time dimension based views (duplicated 4 times for time1/2/3/4 dimensions).
If you need to use 2 out of 4 time dimensions in your query (by applying an application based filter/restriction) then you would need to alias the relevant view(s) 2 times and get your result.
If you needed to use all 4 time attribute dimensions, then you need to alias the time dimension view 4 times.
12 views represent the full spectrum of possible report/query usage scenarios available to users via these 4 time dimension views. A typical session with various reports/queries is unlikely to refer to all of them (possible but not likely).
In different scenarios, any one of the 3 views - dimension view, h1, h2 view - comes into play for each dimension.
In every query/report scenario, only 1 out of the 3 applicable views would be needed or suitable based on the query/report specifics.
E.g:
* If your query is at lowest level and is also referencing attributes/descriptions which pertain to >1 hierarchies, then the view v_time<N>d is the appropriate view to use. Neither vtime<N>h1 or vtime<N>_h2 will be able to give you the results you require if you have attributes defined exclusively in either hierarchy h1 or h2.
* Setting the default hierarchy to either h1 or h2 determines which hierarchy view is expected to be used/preferred in cases where dimension is not referenced or where Top level of dimension is desired in query/report. E.g: Top level is same for both hierarchies h1 and h2 but read the Top Level values from h1 view as h1 is marked as the default hierarchy for dimension.
* The level of the query is generally determined by the lowest level of the attributes/fields in your query. If you have attributes like Long_Desc, Short_Desc (available at all levels/all hiers) and you have 2 attributes from Hierarchy H1 and from levels L1 (one level above leaf) and L3 (higher than L1 by 2 levels)... then the query uses v_time<N>h1 and will apply a filter to choose/constraint levelname field = 'L1'.
* Reporting tools like OBIEE etc. have security filters defined to ensure that the joins to missing dimensions and the appropriate dimension level filters appended to the query in a transparent "behind the scenes" manner.
NOTE: By modeling the time based fields as different dimensions, you get the benefit of time series analysis along any of the 4 dimensions. You can create a Month-to-Date Revenue measure along "Show Biz Time" dimension and also create a Year-to-Date Revenue Measure based on "Booking Time" dimension. Also compare % variance of "MTD - Revenue - Show Biz Time" with corresponding value for Last year "MTD LY - Revenue - Show Biz Time". This is the benefit you get by breaking out the 4 columns in relational fact and modeling them as explicit time dimensions. You pay a small price in terms of cube space expansion, time dimension load increase but as acknowledged earlier, this dimension can be loaded incrementally, the full load process is done only initially. It is an offline process (batch) and does not hamper query performance.
Regd
3) In the underlying star schema, we have only one time dimension table, in the ticket booking fact table, we have 4 dimension columns referencing the same time dimension table. So it's intuitive to expect similar design in the cube. Current approach seems like replicating the dimension table for each dimension column.Do you have parentage along h1/h2 laid out in terms of different parent columns ... 06:30am should have parent D1 in h1 hierarchy (regular day) and parent=D1-1 in h2 hierarchy (business day)?
h1=regular day hierarchy
h2=biz day hierarchy
In that case, perhaps you want "Booking Time" (Time1) to use view: v_time1_h1 by default and "Booking Biz Time" (Time3) to use v_time3_h2 by default.
Using h2 hierarchy for Time1 or h1 hierarchy for Time3 would be illogical/invalid. Maybe you can simplify things by removing the hierarchies from olap model which dont apply to the dimension.
HTH
Shankar

One Dimension Table Deletion issue

Dear Guru's
Query run's very long time to execute in prod system
In our production system, we have a Cube in which one dimension(Eg:- /BIC/PC_17 which show 38% when i run report SAP_INFOCUBE_DESIGNS ). Due to the performance issue of the Query, i am thinking to delete data from SE11 only for this dim table...What will happen with the Factable?
Since Dim table has reference to Factable...Is it correct if i delete the entries for this Dimension table for the perfomance issue?
PLease suggest me i am very much confused, also this is a prod system....so pls give me only good answers..no gess please...
Also ....suppose if this Dim17 is not used in Query Does it make sense to delete entries from this Dim 17 for Query performance, since this exist's in cube...
Thanks in adv,
Dev

Hi,
you should not do that. In no case. There are some ways to speed up a query. Use them: e.g. building aggregates, running rsrv to check the cube and possibly have unused entries of dimensions deleted, compress the cube....
regards
Siggi

One dimension used twice in a cube

Similar Messages

Maybe you are looking for