Understanding Aggregation for Hierarchies
Hi:
OBIEE 11.1.1.6.5
I'm trying to understand aggregation in the context of a dimension's hierarchy. When I create a level-based hierarchy for a dimension there are two places to set the aggregation level. The first is on the source for the dimension table, the other is the source for the joined fact table. My question is, what is/are the purposes for these two options? When would I set one over the other? Currently, it seems to get the aggregation to work, I'm setting the level on the dimension table source and not the fact source. Is this the correct way?
Any insights are greatly appreciated.
Thanks.
Hi,
The purpose of setting the aggregation levels on dim/fact is to let BI server know which LTS to pickup for the user's requested report criteria.
If you have one aggregated table that has data at month level and another base table that has data at day level.
In this case, you will set the content level for time hierarchy as month_ level (since that is the lowest level of granularity present in the table) for the aggregate table
and then you will set the content level for time hierarchy as DAY_ level (since that is the lowest level of granularity present in the table) for the base table.
If you run a report quarterly, BI server looks in both LTSes above and chooses the aggregated table which will give you better performance.
Thanks,
Rajesh Gurram
Similar Messages
-
Configuring measures with custom aggregation for hierarchies
Can I Configuring different measures with different aggregation for hierarchies . in AWM the aggregation for measures are disabled, am I missing something.
For example
Dimension : supplier
Measure : risk score
Aggregation : maximum
Dimension : supplier
Measure : spent
Aggregation : sumYes it is possible. But make sure that you have turned the compression off in the cube
-
Values not getting displayed in the selection screen for Hierarchies?
Hi All,
Our system was upgraded from BW 3.1 to 2004s. The quality system displayes values when query result is filtered for a hierarchy infoobject but the production system does not display any values and sometimes throws an error 'The operation failed because of a server error. See system logs.' There is no changes done for hierarchies in both the systems.
Please let me know if anyone has had similar problems.
Thanks,
KrishnaHi All,
Our system was upgraded from BW 3.1 to 2004s. The quality system displayes values when query result is filtered for a hierarchy infoobject but the production system does not display any values and sometimes throws an error 'The operation failed because of a server error. See system logs.' There is no changes done for hierarchies in both the systems.
Please let me know if anyone has had similar problems.
Thanks,
Krishna -
How to prevent BPC from automated calculating for hierarchies / nodes?
Hi experts,
I am looking for a practicable way to prevent the system from automated calculating for hierarchies, and especially for nodes within hierarchies?
Let's say, I have ENTITIES (AS A HIERACHY) in rows and ACCOUNTS in COLUMNS. Now I want the system to block adding up the values for one specific account on node XY. Instead of the sum of all base member entities, the cell for account XY should be left empty.
Is there a practicable way to deal with this?Hi Stefan,
you can prevent the system to calculate a node by editing the Formula Property in the Dimension.
You can insert 0 or null to the Formula field for the specific element.
But in this case the values from your leaf elements lying under your node wont be calculated even in higher hierarchy nodes.
For ex.: with the following structure
-A
---A1
A11
A12
---B1
B11
B12
If you set A1 to 0, the top node A will only be calculated by values of B1
Regards
Jörg -
Steps to load the data by using flat file for hierarchies in BI 7.0
Hi Gurus,
steps to load the data by using flat file for hierarchies in BI 7.0hi ,
u will get the steps int he following blog by Prakash Bagali
Hierarchy Upload from Flat files
regards,
Rathy -
Best practice for hierarchical DTOs?
Hi!
Can someone tell me the best practice for hierarchical DTOs?
Use case: I've got a User object which holds one Folder object, which in turn holds a Set of Folder objects (children).
class User {
Folder rootFolder;
class Folder {
Set children;
}Normally, I'd fetch the user data with the help of a DAO from the database, copy the requested properties into a User-DTO, which will be transfered to the view. But what about the Folder objects?
Should I create a Folder-DTO class and copy each Folder property into the respective DTO object - in other words: rebuild the whole hierarchy? Or is there a better solution?
Thanks a lot!
WalterNormally, I'd fetch the user data with the help of a
DAO from the database, copy the requested properties
into a User-DTO, which will be transfered to the view.
But what about the Folder objects?
Should I create a Folder-DTO class and copy each
Folder property into the respective DTO object - in
other words: rebuild the whole hierarchy? Or is there
a better solution?It isn't recursive right?
So this is just a standard association.
How you handle it depends on usage.
As a guess perhaps you are thinking that you can only have one User DTO class. That isn't true. You can have several. For example one that contains the association and one that does not.
Or you have just one. And either it contains the association or you provide another mechanism that returns just the association given a specific instance of a User DTO (or some other identifier to the specific User.) -
Functional specs for Hierarchies
Hello GURUS,
Please any one provide me functionalspecs and Technical specs for Hierarchies.
Thanks,
Sekhar.Hi Utpal,
Welcome to SDN,
Being a newcomer to forum before posting the thread read the forum rules,you are asking to send the details to your mail which is forbidden by the forum.
Coming to the thread Customer Hierarchies stored in table KNVH,go to SE16N give the table KNVH get valid inputs and build
valid logic.
Regards
Ram -
Open Hub extraction for Hierarchies
Hello,
I need to create a Open Hub on the 0MAST_CCTR in HR-PA. Unfortunately the Open Hub does not give you the option for hierarchies. Any ideas?
Thanks,Hi,
If it is a flat file output then you can use the below program mentioned in SAP how to document. We are using this in our system.
http://www.sdn.sap.com/irj/scn/go/portal/prtroot/docs/library/uuid/0403a990-0201-0010-38b3-e1fc442848cb?quicklink=index&overridelayout=true
Regards,
Raghavendra. -
Setting "Aggregation For Plan" for Scenario members in Dimension Library
In the Dimension Library, I have a Scenario which has two members (B and C) which roll up to a parent member (A). Unlike the other types of dimensions, I am unable set the Aggregation as the properties (Aggregation for Plan 1, Aggregation for Plan 2, Aggregation for Plan 3) are not available under the Planning category.
If I view the outline in EAS, I am able to modify the aggregation for the A and B members directly but I'm unsure as to what impact this would have on the planning application.
Any ideas why the properties aren't available? The BPMA Administrators guide seems to suggest I should be able to do this.
Can I go ahead and modify the members direct in EAS?
Cheers.I did take a look at the custom dimension and it's members with their storage properties both in DEV and PROD and there were NO Dynamic/Label Only members. Checked the relational database as well and it too showed that there were no members with Dynamic/Label Only storage properties in the custom dimension.
To my surprise this morning, when I checked the numbers in EAS the two columns had the same numbers. It looks like its fixed now as the IMPLIED_SHARE setting took effect and both the "Members in Dimension" = "Members Stored".
I noticed that the server team had stopped and started all the server again yesterday night sometime in DEV so maybe that's when the setting actually took affect and for some reason it didnt take affect the first time they restarted the servers Or maybe because I refreshed the database again from workspace this morning. I am not sure which one fixed it.
Thanks Celvin for your time and inputs.
- Krrish.
Edited by: Krrish on Apr 25, 2013 9:39 AM -
Different aggregation for different Dimensions
Hello,
is it possible to have different aggregations on different dimensions.
I have following situation:
I have a measure per client and day.
I'm interested in the maximum per month from the daily sums over clients.
In the measure properties I can only choose between Maximum and Sum in general but not per Dimensions.
To clearify what i mean here is some sample data.
* * Client A * Client B *
* 2014-11-28 * 7 * 8 * SUM() = 15
* 2014-11-29 * 6 * 8 * SUM() = 14
* 2014-11-30 * 6 * 10 * SUM() = 16 <-- monthly max
* 2014-12-01 * 7 * 8 * SUM() = 15
* 2014-12-02 * 5 * 12 * SUM() = 17 <-- monthly max
* 2014-12-03 * 6 * 9 * SUM() = 15
This data is stored in my fact table with reference to date and client dimensions.
This example data would have to be reported as:
/* Report on measure
* * Measure *
* 2014-11 * 16 *
* 2014-12 * 16 *
* Report on measure per client
(max per client and month)
* * Client A * Client B *
* 2014-11 * 7 * 8 *
* 2014-12 * 7 * 12 *
Can this be achieved with SSAS? Didn't find any property for that on the measure.
Best Regards,
ThomasHi Thomas,
According to your description, you want to calculate different aggregation for different dimensions, right?
Based on your scenario, I tested it on AdventureWorks cube, the query below is for you reference.
with member [Customer].[Country].[USA & Canada] as
Aggregate( { [Customer].[Country].&[United States],
[Customer].[Country].&[Canada]
member [Measures].[MaxAmount]
as
max([Date].[Calendar].currentmember.children,[Measures].[Internet Sales Amount])
select {[Customer].[Country].&[United States],[Customer].[Country].&[Canada],[Customer].[Country].[USA & Canada]} on 0,
[Date].[Calendar].[Month].members on 1
from
[Adventure Works]
where [Measures].[MaxAmount]
Here is similar thread with yours, please see:
https://social.technet.microsoft.com/Forums/en-US/1bd493ef-f957-4fd5-916b-ee60639106c3/calculated-member-different-aggregations-on-different-dimensions?forum=sqlanalysisservices
Regards,
Charlie Liao
If you have any feedback on our support, please click
here.
Charlie Liao
TechNet Community Support -
Transfer Rule Inactive for Hierarchies
Hi All
I have a Typical Problem. i have activated some datasources (3.5) for Hierarchies. we have total four of them. these are working perfectly fine in Dev and we have successfully transported them to QA without issues. But wen we transport the same to Prod one of the transfer rule is giving an error. when we transport only this transfer rule and the corresponding DS, it becomes active, but another one goes inactive. if we transport all the four again, first one again becomes inactive and the other three are active.
eg : i have four transfer rules A, B , C and D. When i transport them A is active and the other 3 are active. so i transported only A with the DS and the communication structure, C becomes inactive. so we transported all the four again. now A becomes inactive. the Transport requests (logs) do not give any error.
we are using Bi 7.0 System.
Has anyone faces this issue. and help me out to rectify the same.
Regards
DVicky,
Can you please elaborate on "it is not letting me change"? What error do you get? This should not happen in development system.
May be someone is keeping the object locked. Highly likely that you are keeping it locked in a different session. Close all the sessions and try again.
If the object is locked and you want to forcibly unlock it, use transaction SM12. -
Hi,
i use in my query different hierarchie nodes in columns
For example
COL A: Hierarchie Countries Node: World
COL B: Hierarchie Countries Node: Europe
in an other Query
COL A: Hierarchie Countries Node: World
COL B: Hierarchie Countries Node: Asia
So i'd like to use a variable for these nodes to enter Europe or Asia.
I've a variable for hierarchies, but when i use it i can only enter the name of the hierarchie. Here countries.
How can i enter nodes in a variable and store them in one workbook with different tables. Both tables use the same query but one with Europe and one with Asia
Thanks for help .... MarkusMarkus,
Do you mean to say :
You have different hiearchies in two different queries and would like to enter only the node ? did you try using Hierarchy node variables - this should allow you to select a hierarchy node from the hierarchy attached.
or did you mean that depending on your node input the query should select the appropriate hierarchy - you can try this using a user exit not sure if it can be done in any other way.
Arun -
Hi Experts,
Is there a way to set Exception aggregation for a restricted key figure?
Thanks
GMSHi Prasad,
A restricted key figure is a normal key figure, so u can set the exception aggregation in the Bex Options in the Key figure properties.
Thanks,
Krish -
I cannot change prefix for hierarchies in dimension properties, when I select some Hierarchy, change prefix to some other letter(s), click update, and reselect Hierarchy to view it updated, i always get dimension prefix !
This happens also during creation of dimension, no matter what I write in prefix during hierarchies definition, in created dimension it shows dimension prefix instead...
Changing name and description of hierarchy works.
OWB is 9.2.0.3, this also happened on base 9.2.
In 9.0.4 it was working properly.
Please help...Hi Piotr,
We have identified as a problem as well and are investigating this. It is logged under bug 3378279, which you should be able to track in metalink.
Thanks,
Jean-Pierre -
Hi all,
why and when we create aggregation for MPOS. can anybody explain.. Thans in advance..Hi, Chandra,
The aggregation help to improve the performance by storing frequently accessed information at aggregate level.
The data will always be stored at the detailed level in the planning area. If you need to access aggregate level data very frequently, the data can be stored separately at aggregate level by specifying aggregate in MPOS.
So, in this way the access to aggregate data will be fastest since system doesn't need to do any data aggregation from detailed level.
Working with aggregate data will need more memory!
Thanks,
Regards, Marius
Maybe you are looking for
-
How to set up recordings of audio and video so that they will sync
I'm editing footage of a dance concert that used CDs for some of the dances and a live pianist for the remainder. In addition to the audio captured by the camera, I have all the CDs that were used and an additional audio recording of the piano that s
-
I have hp 430 laptop,how can i re install the windows 7, if i not have original windows on my laptop. i have oem number, But i have troubol to finishing installation of window.
-
Travel Expense - Posting run error - Error during creation of posting run
Hi Guru's, Look forward for your insights. Here you go with my question. We are implementing Expense Mgt with Portal with out scope for payroll in SAP. I have set up Expense type, WT, Symbolic A/c to G/L Accounts. Im performing Unit testing currentl
-
Hi, i am connecting a Bluetooth device to my laptop. I notice that Message Analyzer could capture Bluetooth messages only if I am using windows 8? I am using window 7. What driver or software patches do i need to install in my windows 7 laptop, in or
-
How to truncate a field in a Select Query
Hello Gurus, I am new in SAP, i have a question i am making an appointment letter in SMARTFORMS , in global declaration (initialization) i am using a select query SELECT SINGLE BET01 FROM PA0008 INTO G_BET01 WHERE PERNR = WA_PERNR. where BET01 is the