Essbase's attribute dimensions in OBI 10.3.4.1

Hello everybody:
As far as I know, it is possible to use Essbase attribute dimensions in OBI 10.3.4.1. But, in your opinion, is there any problem when integrating them or they work in OBI as standard dimensions?. Have somebody had any issue with this integration?
Thank you in advance
Regards,
Javier

Hello:
I have read this article before writing this post. But it refers to Essbase 9.3.1 (I refer to 11.1.1.3) and OBI 10.1.3.3.2 (I refer to 10.3.4.1). One of the goals of my questions is knowing if something has changed since that article was written, because it implies some manual changes in the rpd.
Thank you
Regards
Javier
Edited by: Javi M on 28-feb-2011 4:51

Similar Messages

Deploying Essbase Text-Attribute Dimensions

Hello,
we have an Essbase-cube with a text-attribute dimension. All associations are fine, also the validation. The deployment fails with the error:
"Level zero members defined as dynamic calc must have a formula assigned to them"
As it is an attribute dimension, there is something wrong.
Any help apreciated!
Michael

Are you sure you don't have an Account or some other dimension's member that has been tagged as dynamic calc and yet does not have a formula?
That's the message you'll get when that happens.
The Planning refresh should (hopefully) show you which member is that way. Failing that you can go into Essbase and try doing outline validations -- I suspect the error will not have gone across but it may be worth a try. If the latter is the case, you are going to have to compare the two -- Planning and Essbase to find that dynamically calculated member without a formula.
Regards,
Cameron Lackpour
P.S. The message sort of makes sense, sort of -- attribute dimensions are dynamic, but I've never personally seen this error.

Multiple Essbase-derived measure dimensions in OBI EE

Hi guys,
I have a requirement here which basically requires the usage of measures coming from two different measure dimensions of Essbase (Account and Scenario) next to each other in reports.
I set both to dim type "measure dimension" and made sure the respective measures exist as physical cube columns. Then I tried two alternatives:
1.) One logical fact table - result spool:
-------------------- Logical Request (before navigation):
RqList distinct
Actual:{DAggr(Account.Actual by { } )} as c1 GB,
Actual LY:{DAggr(Account.Actual LY by { } )} as c2 GB,
Target:{DAggr(Account.Target by { } )} as c3 GB,
Vs Actual LY:{DAggr(Account.Vs Actual LY by { } )} as c4 GB,
Vs Actual LY %:{DAggr(Account.Vs Actual LY % by { } )} as c5 GB,
Ops:{DAggr(Account.Ops by { } )} as c6 GB,
Sales_In:{DAggr(Account.Sales_In by { } )} as c7 GB
+++Administrator:2a0000:2a000a:----2008/11/10 14:29:55
-------------------- Execution plan:
RqList <<60982>> {for database 0:0,0} distinct
D1.c1 as c1 {for database 0:0,0},
D1.c2 as c2 {for database 0:0,0},
D1.c3 as c3 {for database 0:0,0},
D1.c4 as c4 {for database 0:0,0},
D1.c5 as c5 {for database 0:0,0},
D1.c6 as c6 {for database 0:0,0},
D1.c7 as c7 {for database 0:0,0}
Child Nodes (RqJoinSpec): <<61009>> {for database 0:0,0}
RqList <<60949>> {for database 3023:179879:Full cube,34}
AggrExternal(INSIT.Actual) as c1 GB {for database 3023:179879,34},
AggrExternal(INSIT.Actual LY) as c2 GB {for database 3023:179879,34},
AggrExternal(INSIT.Target) as c3 GB {for database 3023:179879,34},
AggrExternal(INSIT.Vs Actual LY) as c4 GB {for database 3023:179879,34},
AggrExternal(INSIT.Vs Actual LY %) as c5 GB {for database 3023:179879,34},
AggrExternal(INSIT.Ops) as c6 GB {for database 3023:179879,34},
AggrExternal(INSIT.Sales_In) as c7 GB {for database 3023:179879,34}
Child Nodes (RqJoinSpec): <<60952>> {for database 3023:179879:Full cube,34}
INSIT T179892
) as D1
+++Administrator:2a0000:2a000a:----2008/11/10 14:29:55
-------------------- Sending query to database named Full cube (id: <<60949>>):
select
{ {Scenario}.{Actual},
{Scenario}.{Actual LY},
{Scenario}.{Target},
{Scenario}.{Vs Actual LY},
{Scenario}.{Vs Actual LY %},
{Scenario}.{Ops},
{Scenario}.{30001}
} on columns
from {INSIT_C.INSIT}
2.) Two logical fact tables - result spool:
+++Administrator:2a0000:2a0007:----2008/11/10 14:49:23
-------------------- Logical Request (before navigation):
RqList distinct
Account.Sales as c1 GB,
Account.Backlog as c2 GB,
Account.Sales_In as c3 GB,
Scenario.Act vs Bud % as c4 GB,
Scenario.Act vs Budget as c5 GB
OrderBy: c1 asc, c2 asc, c3 asc, c4 asc, c5 asc
+++Administrator:2a0000:2a0007:----2008/11/10 14:49:23
-------------------- Query Status: Query Failed: {nQSError: 15018} Incorrectly defined logical table source (for fact table Account) does not contain mapping for {Scenario.Act vs Bud %, Scenario.Act vs Budget}.
+++Administrator:2a0000:2a0009:----2008/11/10 14:49:28
-------------------- Logical Request (before navigation):
RqList distinct
Account.Sales as c1 GB,
Account.Backlog as c2 GB,
Account.Sales_In as c3 GB
OrderBy: c1 asc, c2 asc, c3 asc
+++Administrator:2a0000:2a0009:----2008/11/10 14:49:28
-------------------- Execution plan:
RqList <<37039>> {for database 0:0,0} distinct
D1.c3 as c1 GB {for database 0:0,0},
D1.c2 as c2 GB {for database 0:0,0},
D1.c1 as c3 GB {for database 0:0,0}
Child Nodes (RqJoinSpec): <<37047>> {for database 0:0,0}
RqList <<37051>> {for database 3023:179879:Full cube,34}
INSIT.Sales_In as c1 GB {for database 3023:179879,34},
INSIT.Backlog as c2 GB {for database 3023:179879,34},
INSIT.Sales as c3 GB {for database 3023:179879,34}
Child Nodes (RqJoinSpec): <<37059>> {for database 3023:179879:Full cube,34}
INSIT T179892
) as D1
OrderBy: c1 asc, c2 asc, c3 asc {for database 0:0,0}
+++Administrator:2a0000:2a0009:----2008/11/10 14:49:29
-------------------- Sending query to database named Full cube (id: <<37051>>):
select
{ {Scenario}.{Sales}{Scenario}.{Backlog}{Scenario}.{30001}
} on columns
from {INSIT_C.INSIT}
+++Administrator:2a0000:2a0009:----2008/11/10 14:49:29
-------------------- Query Status: Query Failed: Essbase Error: Unknown Member Scenario.Sales used in query
So basically neither of the two options work at all. Most of the time he just assumes that all that measures are coming from the last hierarchy I defined as a measure dimension...hence the unknown member error.
Likewise with the two logical tables, he screws the query up completely and doens't even manage to resolve the column mappung against the logocal tables.
Has anyone ever tried to mix measure dimensions retrieved from Essbase? Using OBI EE it's very tricky to define the acutal measure containing data points in Essbase.
Cheers,
C.
Edited by: Christian Berg on Nov 10, 2008 2:58 PM
Edited by: Christian Berg on Nov 10, 2008 3:00 PM
Sorry for the reformatting...the forum doesn't like the square brackets in the BI server logs...

Solved. Takes some nibbling around in Answers, but doable with unioning and pivot tables.
Cheers,
C.

Attribute Dimensions and Excel Retrieves

Dear All,<BR><BR>I have a cube that has a sparse dimension with attributes attached (all at the same level). When I retrieve using different attributes in the retrieve the results are not consistent i.e. some of the sparse members with the same attribute are left out of the result. It appears to happen when branches of the dimension in the same area are separated by a branch with an alternate attribute value (from the same attribute dimension). I have cured this by moving the affected branches next to each other, is this something that I need to watch out for on the design side i.e. to do with the way Essbase handles attribute dimensions? Any guidance would be appreciated.

Graham,
This is definitely a supported feature in EIS/9.3.1/ASO. I have many models with this type of structure. How you set it up can vary. Usually my Attribute Hierarchies are not that deep, only two to three levels, maybe four in a rare case, so I don't usually use a parent child table to set up the hierarchy (I'm not saying that it won't work, it might, I haven't tried, but same steps should apply). In a typical model I will have my stock table which has a buyer field. Then in another table I will have my attribute structure which will have columns for buyer, teams, and categories.
In EIS OLAP model, you add your attribute hierarchy table and use a join to link it to the main stock table, joining on the buyer field (you are now going from a "star" schema to a "snowflake". Go into the properties and make sure you define all the columns as "Attributes".
Then in Metadata model, drag your categories attribute onto the outline, then drag the teams and set it as a child of categories and finally drag buyer and set as child of teams. You only set the attribute association for the buyer back to the base dimension.
When you run your dim build it will set up your attribute dimension correctly.
Some things to keep in mind, make sure you have a process that ensures for every stock code you have in the main table, you have a matching one in your attribute dim table.
Sometimes, depending on how much manipulation I need to do, instead of joining the tables in EIS I will go back to relational source and create a view that joins the two tables together, then in my OLAP model, I have one table that has three attribute columns, one column for the buyer and then the other two for team and category, from that point setting up in metadata model is the same.
Good luck, let me know if you run into trouble.

How to create an attribute dimension in Essbase Studio?

I cannot figure out how to create an attribute dimension using Essbase Studio. The online help states that I need to "Find the Attribute settings grouping in the center" of the member properties dialog box. I suspect that I'm having a problem figuring this out because I don't have the attribute dimension properly defined in the relational data source or in the mini-schema. I've tried creating a table with no key relationship to the other tables and with no logical joins defined, but that doesn't seem to help. Any ideas?

so in the 11.1.1 version you would click on the member name in the hierarchy you added to the Essbase model and in the center pane of the general tab should be the attributes settings section.
If I had a dimension called stores wth two groupings unser it region and store_type, I would click on store_type and add the attribute info

Hierarchical attribute dimension with Essbase Studio

Hi everybody,
I need to load product dimension which is organized in a relational parent-child table like this:
(parent_code, member_code, member_alias, brand, consolidation, formula)
Every product has a brand and I need to load brand as attribute dimension. The thing is Brand is not flat but it has its own hierarchy. For Brand I have another relational table where data is organized parent-child like this:
(brand_parent, brand_child, brand_alias).
I need to do this load with Essbase Studio but I do not known how to create the Product hierarchy in cube schema in order to load the hierarchical parent-child attributes. Documentation and tutorials only show examples for flat attributes. I thought of these two ways but not sure if will work:
case 1) Product hierarchy like this:
Parent_code
Member_code
Brand_parent
Brand_child
Member_code
case 2) Product hierarchy like this:
Parent_code
Member_code
Brand_parent
Brand
Member_code
Has anyone had to this? How did you do it?
Any help will be very appreciate.
Thank you,
Daniela

I've not tried it, but you should be able to build the attribute hierarchy within the product hierarchy. sort of like
Parent Code
---member Code
Brand Parent
---Brand Code
-----Member Code.
A couple of things
1. Remember to create the self join between Parent_code and Member code
2. IF this does not work, I suggest creating either a user defined table or a view that just gives you distinct brand parent and Brand Code. Join the Brand Code to the product table on brand_code = brand code and use that table to build that hierarchy.

Attribute Dimension Essbase

Dears,
I would like to know how Attribute Dimension Essbase working in OBIEE,
I'm Using ASO & i import Time Dimension and Attribute Dimension for Time (Sun, Mon, etc).
I want to know how its working in Report.
I want to See the Time dimension with This Day name ?
Is it possible ?
Thanks in advance.

Any Help please ?

Essbase Attribute Dimension

When we retrieve totals on base dimension without specifing the attributes the numbers turn out right. HOWEVER when we retrieve specifying the attributes the totals consolidate everything inspite of having tilde on some of the members.

You are correct, when you retrieve and include an attribute dimension, then it will add everything regardless of the sign. eg. profit = revenue less expenses, however, throw an attribute in there and then suddenly, profit will be revenue plus expenses. Wrong.This is because attributes are always dynamically calced and they will only sum, count, min, max or avg. This means that you can't have attributes associated to dimensions that roll up with anything other than pluses(+). To make it more complicated, if you do have attributes against a standard roll-up dimension and the other dimensions have a dynamic calc member then the dynamic calcs will be a sum again regardless of the sign. So the other dimensions have to be stored values.

'Essbase Error: Repeated dimension [Accounts] in MDX query'

Dear Essbase Gurus,
I have an error in a query passed to my Essbase cube from an OBI front end, but the errors states quite clearly that it is an essbase error:
'Essbase Error: Repeated dimension Accounts in MDX query'
This is the query it generated:
With
set Accounts3 as 'Accounts.Generations(3).members'
set Attribute Calculations2 as '[Attribute Calculations].Generations(2).members'
set Events5 as 'Events.Generations(5).members'
set Month3 as 'Month.Generations(3).members'
set Scenario2 as 'Scenario.Generations(2).members'
set Year2 as 'Year.Generations(2).members'
set Axis1Set as 'crossjoin ({Accounts3},crossjoin ({Attribute Calculations2},crossjoin ({Events5},crossjoin ({Month3},crossjoin ({Scenario2},{Year2})))))'
select
{Measures.Total e-revenue} on columns,
NON EMPTY {Axis1Set} properties ANCESTOR_NAMES, GEN_NUMBER on rows
from Richmond.RichPOC
As an Essbase newbie I'm unfamiliar with mdx and have no idea what this means, any help would be appreciated.
Thanks
Ed
Edited by: Ed Higgins, AltiusConsulting.com on Oct 15, 2008 9:26 AM

Hi garycris,
Thanks for taking a look, I fear you may have hit the nail on the head, as I mention we are reporting on this using OBIEE, so we have imported the cube (BSO) into our OBI repository, OBI flattens the cube to make it look like a star schema it doesn't recognise the Accounts dimension correctly so we have to set it to be of type 'Other' after we have imported it, otherwise none of the figures in the Accounts dimension are actually reported (you can see the members but when included in a query no data is returned).
Another issue we found was that OBIEE appeared to substitute the name 'Measures' for Accounts when building the query, so we made 'Measures' the default alias for the Accounts dimension, to add to the issue I probably should have mentioned that the aggregation rule in OBIEE for this fact column has been set to SUM rather than External_Agg, but Oracle support have previously indicated that this is OK to do. The issue originally arose because we wanted to extend on of our dimensions with additional attributes stored in a SQL database table.
Sorry it is all a bit messy, the whole OBI with Essbase as a source is still relatively new, and it looks like there may be a few things for the mighty O to iron out, I posted this in here because the OBIEE forum has no posts with similar issues, so thanks for your help so far and anything else you can add.
Ed

Error While Building Attribute Dimension in ASO Cube

Hi all,
Cuurently I'm trying to bulit an ASO application in Oracle EPM 11.1.1.3
When I'm building an attribute dimension "QSize" in ASO cube ,I'm Getting some Warnings
I'm using Flat file as my source AND MY RULE FILE IS
GEN2,QSIZE, ALIAS2,QSIZE
1, ">100"
2, "100-249"
3, "250-499"
4, "500-1000"
5, "<1000"
I VALIDATED THE RULE FILES,
when i try to load the data into the essbase,Its giving me a warning stating that
Record #1 - Incorrect Syntax For Member <100 (3305)
*1 "<100"*
and i don't see the alias name for member 1
I came know that Essbase doesn't allow ">" character at the beginning of Dimension Or Member Name
Please let me know if there is Any other way,where I can Assign the alias "<100"
Thx in Advance
Regards,
RSG
Edited by: SG on Jan 22, 2011 3:03 PM
Edited by: SG on Jan 22, 2011 3:04 PM
Edited by: SG on Jan 22, 2011 3:06 PM

Could you get away with an underscore? I think that works. You will have _<100 which is not ideal but might be worth a try.
Failing that, go with GT and LT. <--This is incredibly lame, but sometimes users go for the "computer is too stupid" answer. Which in this case, is sort of true.
Regards,
Cameron Lackpour

How to design Flat file for loading attribute dimension in a planning application

Dear Gurus,
I have a requirement to extract attribute dimensions from an essbase application and load it to another planning application. I have a dimension called Program and two attribute dimensions Sales Manager, Accounts manager associated with Program dimension in Essbase application. I will Extract these dimensions using Essbase outline extractor. After Extracting the attribute dimensions I have to load these dimensions to planning applications using outline load utility. Kindly guide me how to design the flat file for loading attribute dimensions in planning application.
Thanks and Regards
SC

You could dig through the docs and try to figure out the file format manually, or you could do this the easy way. Simply use the Outline Load Utility to export your attribute dimension from Planning. The export file format is the same as the import file format. You might have to manually add a couple of test members to your attribute dimension so that your export file has some content. Then simply update the file you exported, and import it.
(I am assuming you have already manually created the Attribute dimension in Planning, and that you simply need to add members to it.)
Hope this helps,
- Jake

Performance problem with attribute dimensions - please help!

Hi everybody,
I run OBIEE reports on top of EssBase 9.3 ASO cube.
My cube has 8 regular dimensions and 10 attribute dimensions.
I am running a query based on:
- 3 dimensions (TIME, BATCH, MODEL),
- 2 attributes of BATCH ( SHIFT, SHIFT_STATUS )
- and one measure.
The query runs 3 minutes (!) and returns 190 rows.
If I remove the attributes, the query runs 4 seconds for the same number of result rows.
My MDX:
With
set [TIME6] as 'Filter([TIME].Generations(6).members,
([TIME].currentmember.MEMBER_ALIAS = "2009/01" OR [TIME].currentmember.MEMBER_Name = "2009/01"))'
set [TIME7] as 'Generate([TIME6], Descendants([TIME].currentmember, [TIME].Generations(7), leaves))'
set [BATCHES] as '[DIM_BATCH_HEADERS].Generations(5).members'
set [SHIFT2] as '[SHIFT].Generations(2).members'
set [SHIFT_STATUS2] as '[SHIFT_STATUS].Generations(2).members'
set [MODEL2] as 'Filter([MODEL].Generations(2).members, ([MODEL].currentmember.MEMBER_ALIAS = "SNF_PCK" OR [MODEL].currentmember.MEMBER_Name = "SNF_PCK"))'
select
{[Accounts].[PROD_ACTUAL_QTY_KG]
} on columns,
NON EMPTY
{crossjoin ({[TIME7]},
crossjoin ({[BATCHES]},
crossjoin ({[SHIFT2]},
crossjoin ({[SHIFT_STATUS2]},{[MODEL2]}))))}
properties ANCESTOR_NAMES, GEN_NUMBER on rows
from [COP_MAAD.Cop_Maad]
Thank you in advance,
Alex
Edited by: AM_1 on Feb 26, 2009 6:35 AM
Edited by: AM_1 on Feb 26, 2009 7:03 AM

Hi Glenn,
I though that the reason might be the crossjoin with a very large set. The main dimension in this query has ~40000 members.
Is there in EssBase MDX some analog for NONEMPTYCROSSJOIN() function that exists in MSAS?
As these two attributes belong to the dimension that is used in the query, for every member of the dimension, there is one and only one member of each attribute. Therefore I though that Essbase should "know" how to behave.
Thanks for an idea. I will try to increase the space for aggregation - hope this will help...
Best Regards,
Alex

Attribute Dimensions with Integration Services

Good evening, I'm creating an ASO Essbase model through Integration Services v9.3.1 and I am having difficulties with creating an Attribute dimension that contains a hierarchy.
We have a list of stock codes which are grouped by vendor in a dimension, each stock code has an attribute assigned to it for the buyer responsible.
The business requirement is to start from the top level of the attribute dimension, drill in to categories, then to drill in to teams, and finally to the buyer... so we effectively require 3 levels within the attribute dimension.
I have successfully created both the vendor (physical) and buyer (attribute) dimensions through EIS without problems, but when I load to Essbase all of the buyers are added at the top level of the dimension and I can't seem to specify that there should be a parent/child hierarchy to build the attribute dimension from.
Is this possible, and if so any advice you can give as to how I achieve this would be greatly appreciated.
Thanks in advance,
Graham

Graham,
This is definitely a supported feature in EIS/9.3.1/ASO. I have many models with this type of structure. How you set it up can vary. Usually my Attribute Hierarchies are not that deep, only two to three levels, maybe four in a rare case, so I don't usually use a parent child table to set up the hierarchy (I'm not saying that it won't work, it might, I haven't tried, but same steps should apply). In a typical model I will have my stock table which has a buyer field. Then in another table I will have my attribute structure which will have columns for buyer, teams, and categories.
In EIS OLAP model, you add your attribute hierarchy table and use a join to link it to the main stock table, joining on the buyer field (you are now going from a "star" schema to a "snowflake". Go into the properties and make sure you define all the columns as "Attributes".
Then in Metadata model, drag your categories attribute onto the outline, then drag the teams and set it as a child of categories and finally drag buyer and set as child of teams. You only set the attribute association for the buyer back to the base dimension.
When you run your dim build it will set up your attribute dimension correctly.
Some things to keep in mind, make sure you have a process that ensures for every stock code you have in the main table, you have a matching one in your attribute dim table.
Sometimes, depending on how much manipulation I need to do, instead of joining the tables in EIS I will go back to relational source and create a view that joins the two tables together, then in my OLAP model, I have one table that has three attribute columns, one column for the buyer and then the other two for team and category, from that point setting up in metadata model is the same.
Good luck, let me know if you run into trouble.

Business Rule - Can you fix on children in attribute dimension?

HI,
Base Dimension: Service
Attribute Dimension: Service_Code
If I have the following hierarchy in an attribute dimension, how can I fix on all members of the Service dimension that have the attributes Model A, Model B and Model C attached them without listing them individually? Is there a way to fix on children of 'IMH Models' somehow?
Service_Code
   IMH Models
      Model A
      Model B
      Model C
   CAD Models
      Model D
      Model E

Hi
in the FIX use the @ATTRIBUTE("IMH Models"). By default it will fix on all base members associated with the children of the attribute parent you have specified. See below text from Tech Ref
Description
When @ATTRIBUTE is used with a non-level 0 member of an attribute dimension,
Essbase returns all base members that are associated with the children of the
attribute member. For example, in the Sample Basic database,
@ATTRIBUTE(Large) returns all base members that fall into one of
the population ranges for the attribute parent Large.
Thanks
Anthony

Attribute dimension hierarchy

what is use of hierarchy in Attribute dimension. Any one can give example when to have it.

The essbase documentation has a good section on understanding attributes - http://docs.oracle.com/cd/E17236_01/epm.1112/esb_dbag/dattrib.html#dattrib497341
Cheers
John
http://john-goodwin.blogspot.com/

Essbase's attribute dimensions in OBI 10.3.4.1

Similar Messages

Maybe you are looking for