Issues in aggregation

Hi,
I have a requirement as below:
Measure Ver
100 1
150 1
110 2
150 2
50 3
100 3
I have other dimensions as well which I have not included for clarity sake. I have created a cube in OLAP (10g R2) using AWM (10g R2). I have created a dimension on ver as well. My requirement is when I look at the measure when 1 is selected on ver dim, I want to see values 250, when I select 2 I need to see 510 (sum of measure for 1 and 2) and similar for 3 I need to see the values summed for 1,2 and 3.
I created the ver dimension table as below.
lvl1 lvl2 lvl3
1 2 3
2 3
3
then created a dimension with three levels and mapped these columns to the levels. finally in the cube mapped the ver column of the fact table to the three levels in the cube under this dim.
When I run the report and drill down to the different levels the values are the same (i.e) the values corresponding to level1 ('1') is shown and when I drill to 2 it still shows the values corresponding to 1 and does aggregate values for 1 and 2 as required. How can achieve my requirement?
Appreciate any inputs on this.
Thanks a lot in advance!
Maruthi

I think you can also let "ver" remain as a plain dimension (w/o hierarchy, w/o levels) and create a formula CUMMEASURE as below and expose the new formula measure CUMMEASURE instead of the base measure MEASURE.
CUMMEASURE = CUMSUM(MEASURE, ver)
You can define such a formula in your AW with the help of the "Creating OLAP Calculations using Excel" utility found on OTN/Oracle OLAP website.
NOTE: "ver" dimension should be loaded such that the dim members 1, 2, 3 are created in the right order. Start with 1, 1 preceding 2, 1,2 preceding 3 etc.
You may have to verify/ensure this after the AW maintain load process is completed.
LIMIT ver TO ALL
SORT ver A convert(convert(ver, TEXT), INTEGER)
MAINTAIN ver MOVE VALUES(ver) FIRST

Similar Messages

Issue with aggregation

Hi
There is a measure where i had put sum function ex SUM(column name) . But i am expecting my query to be select sum( column name) as c1 where (fact table) order by (fields). When i see the physical query i do not see sum(column name) inthe query which clearly states it is not doing sum on that column. But the logical request sends the query as sum(col name). Am i missing any thing in this regard. Please advice me?
Thanks

Yes, i can see the sum(col name) in the advanced ab sql issued.
I have a column 'x' which is a measure and in the database 'x' is not aggregated. I am trying to do aggregation on the report. For that i have used fx as sum(x) . In the advanced xml as well as logical request it shows as sum(x) but when i try to look at the view log in manage sessions which is the physical sql i just see it as x instead of sum(x). That means it is not aggregating. Due to which the number of records are more as there is no aggregation.
What else can i check? Please advice
Edited by: slek on Oct 29, 2012 8:28 AM

DME thailand issue with aggregation

Hi,
Can any one help me in the below issue?
Relating to DMEE for thailand
Invoice amount in payment record 2 different requiremnt one is when it is local currency it should populate gross amount in Dme,
where as when it foreign currency it should populate the net amount in DME.
To resolve this we have created one element under which 2 nodes for local curency and foregn currency.
Now the problem is with total record. it actually aggregated the amount based on the REFERENCE ID given in the element .. since here 2 atoms are there for foregn and local if i am taking 2 reference ID's and calling both giving different reference ID .. it is giving error ..
can any suggest how to get each total in when it is local currency and when it foregn currency ..
is there any setting with aggregation
thank you for your help..

I guess E123456 Member calculation overrides the consolidation value of its children. (E123 +, E456 +...)

Bex Issue : before aggregation

Hi,
I have a problem in Bex that I thought to solve with the "before aggregration" option in Bex Analyzer but this doesn't seem to be possible.
My query looks initially like this for 1 shipment 1344601 that is linked to 2 deliveries
Shipment Delivery Del.Quantity Counter Del.Quantity/Count
1344601   81209721 41.160 KG     2      20.580 KG
1344601   81209722 14.840 KG     1      14.840 KG
Resultrow .............56.000 KG     3       35.420 KG
This 35.420 KG is the right total delivery quantity for shipment 1344601. Note that counter is calculated via a formula with before aggregation. Also the Del. Quantity is included in a formula with before calculation.
When I now remove the drilldown on the delivery my result is wrong :
Shipment      Del.Quantity Counter Del.Quantity/Count
1344601       56.000 KG    3       18.666.67 KG
Resultrw     56.000 KG    3       18.666.67 KG
The system is dividing 56.000 KG by 3 which leads to an incorrect result of 18.666.67 KG.
==> My problem is that I can't put the Del.Quantity/Count on calculation before aggregation, although Del.Quantity and Counter that are used in the calculation have both calculation before aggregation.
Via a little trick (include the calculation in a new calculated keyfigure where I multiply it by 1) I could put it on calculation before aggregation but then the query execution fails.
Does anyone know a solution for this behaviour as I think lots of you might have faced similar situations and I would like to avoid to solve this in data staging as this doesn't seem unsolvable in Bex to me...
Message was edited by: Wim

Hi,
As per the my knowledge, we can use the properties of Before Aggregation in Bex Analyser.
for ex. when u excute your query, by default the result of the calculated key figure is generated after aggregation. But we can change this property to before aggregation by going to context menu of the calculated key figure for the analyer.
But it is considered as a bad practice in SAP.
If you want to use this option in query designer it not available,
because when we use calculated key figure in formulas it calculate as after aggregation. If you want to change it, we can change at analyser level.
Hope u understood.
Regards,
Vamsi
Assign points if useful.

Require Very Urgent Help on Aggregation Issue. Thanks in advance.

Hi All,
I am new to essbase.
I have got an issue with aggregation in Essbase. I load data at zero level and then when I aggregate using CALC DIM i do not get any value.
The zero level load being:
Budget,Version,Levmbr(Entity,0),Levmbr(Accounts,0),NoRegion,NoLoc,NoMod,Year,Month.
When I use default calc or give Calc Dim for the above no aggregation takes place at the parent level.
Requirement :
Values at Version,Region,Location,Model,Year,Month.Budget Level.
Please advice.
Thanks in advance.
Bal
Edited by: user11091956 on Mar 19, 2010 1:07 AM
Edited by: user11091956 on Mar 19, 2010 1:10 AM

Hi Bal,
If you had loaded without an error , and after that your default calc is resulting in not aggregated values. Then I can imagine only one way, it cannot happend is through your outline consolidations.
Check,if data which is loaded at the members does have IGNORE or ~ as the consolidation
Sandeep Reddy Enti
HCC
http://hyperionconsultancy.com/

Aggregation script is taking long time - need help on optimization

Hi All,
Currently we are working to build a BSO solution (version 11.1.2.2) for a customer where we are facing performance issue in aggregating the database. The most common activity of the solution will be to generate data on different scenario from Actual and Budget (Actual Vs Budget difference data in one scenario) and to be used for reporting purpose mainly.
We are aggregating the data to top level using AGG command for Sparse dimensions. While doing this activity, we found that it is creating a lot of page files and thereby filling up the present physical memory of the drive (to the tune of 70GB). Moreover it is taking a long time to aggregate. The no. of stored members that is present are as follows:
Dimension - Type - Stored member (Total members)
Account - Dense- 1597 (1845)
Period - Dense - 13 (19)
Year - Sparse - 11 (12)
Version - Sparse - 2 (2)
CV - Sparse- 5 (6)
Scenario - Sparse - 94 (102)
EV - Sparse - 120 (122)
FC - Sparse- 118 (121)
CP - Sparse - 1887 (2049)
M1 - Sparse - 4873 (4874)
Entity - Sparse - 12020 (32349) - Includes two alternate hierarchies for rolling up the data
The other properties are as follows:
Index Cache - 152000
Data File Cache - 32768
Data cache - 153600
ACR = 0.65
We are using Buffered I/O
The level 0 datafile is about 3 GB.( 2 year budget and 1 year 2 months Actuals data)
Customer is going to use SmartView to retrieve the data and having Planning Plus License only. So could not go for an ASO solution. We could not reduce the members of huge Sparse dimensions M1 and CP as well. To improve the data retrieval time, we had to make upper level members as stored which resolved data retrieval issue
I am seeking for help on the following:
1. How can we optimize the time taken? Currently each dimension is taking about an hour to aggregate. Calc Dim is taking even longer time. Hence opted for AGG
2. Will change of dense and sparse setting help our cause? ACR is ona lower side. Please note that most calculations are either on Period dimensions or FC. There is no such calculation on Account dimension
3. Will change of a few non-level 0 members from store to dynamic-calc help? Will this slow down calculations in the cube?
4. What should be the best performance order for this cube?
Appreciate your help in these regard,
Regards,
Sukhamoy

Please provide following information
1) Block size and other statistic
2) Aggreagation script
>>Index Cache - 152000
>>Data File Cache - 32768
>>Data cache - 153600
Try this settings
Index Cache - 1120000
Data cache - 3153600

POS Sales Posting Aggregated always in Base UOM Why?

Hi All
We have an issue in Aggregated Sales Idoc Posting. There is an article which is having Base UOM as "EA" . It has "CAR" = 5 EA.
so 2 UOMs are there. From POS when the file is posted with 2 CAR , in SAP Billing and Article documents created only with "EA" and not with"CAR"
We faced the similar kind of issue in "Receipt based Sales " and we asked the POS guys to pass the EAN instead on Article Code
They passed the Qualifier "EANN" and the Barcode so that posting is ok for Receipt based sales .We have done this because there is no "SALES UOM" field in Receipt based Message type" WPUBON"
But in Aggregated "WPUUMS" ,Sales UOM field is there and we checked the Idoc and it contains "CAR" as Sales UOM , but while posting ,it takes the Bases UOM. In this case "EA".
Can anyone clarify us whether standard SAP works like this? And is there in any way to make the system to post at other UOM ?
Please help us in this
Regards
Anis

I am also getting the same issue. The right sales UOM is not getting posted to article and Billing documents from WPUUMS01 IDOC in aggreated sales posting. Can anyone help Plz...

Accessing same data in Multiple Excel Worksheets

I am trying to create a summary page based on data that is on subsequent worksheets via the Excel template in Bi Publisher 11g.
I wanted to know if there is a way to reference a Group. So if I had a Department (G_D) and a Employees (G_E) groups in the data model and then I wanted to reference G_D on both sheet 1 and Sheet 2 in the spreadsheet how would I do it.
I created the equivalent in the XDO_METADATa Sheet:
XDO_GROUP_?G_D1? <?(.//G_D)?>
XDO_?G_DNO1? <?(.//DEPTNO)?>
and the Template viewer ran but did not print out any departments. When I remove the XDO_GROUP_?G_D1? it prints out all of the departments in one row, rather than in columns as I would like.

Again, as I got no help from the forum, I resolved this by changing the data model within BI Publisher and created another query for the summary data and then referenced this on the other page.
Will leave thread open for a few days in case someone has a better way of doing it.
I wanted to reference the existing data because:
a) it is already there, why should I have to run another query
b) I am guaranteed to get the same results and won't have any rounding type issues when aggregating the individual values.

Dimension member values not showing (just empty cells)

Hi,
I have an organization dimension (Business > Business Unit> Department > Account) and a single fact (Dollars) on my test repository. When I add my fact and any dimension level to an answers report I can see for example two empty cells for each member at the "Business" level and the result of my fact calculation. I've tried many ways to attempt solving this issue from aggregating my data to compressing all the dimension in a single table and still haven't had success. I also turned on logging and found out that the physical layer SQL being generated returns the correct values when I run it in my query tool.
Has anyone experienced similar behavior? I'm not sure if I skipped a step on my modeling or this is a bug in my OBIEE installation
Thanks a lot!!
-Ignacio

Had this problem when source datatype was nvarchar and the field was coming over as varchar with a length of 0 in the physical layer. Check your physical table definitiion to see the length of the fields that are empty for you, may have to change it to an appropriate value

Wrong Sales UOM in Article and Billing document

Hi,
We have an issue in Aggregated Sales Idoc Posting. There is an article which is having Base UOM as "EA" . It has "CAR" = 12 EA.
So 2 UOMs are there. From POS when the file is posted with 1 CAR , in SAP Billing and Article documents created only with "EA" and not with"CAR".
In Aggregated "WPUUMS" IDOC,Sales UOM field is there and we checked the Idoc and it contains "CT" for CAR as Sales UOM , but while posting ,it takes the Bases UOM. In this case "EA". Its posting the right quantity as 1 for CAR.
Please help

Hi,
There is no any option to change currancy at time of invoice creation
But round the way solution is available
Create sales order with CLP Currancy (VA01) >>> Delivery (VL01N) >>> Create invoice with CLP currancy (VF01)
Then go to t-code EWUO
And change the currancy of sales document
Kapil

OBI 10g merge 2 time dimensions on the same table/graph in analysis

Hi,
My problem is: I have 2 different analyses (working fine) (using OBI 10g), they both use a different time dimension:
time_dim_1 | fact1
time_dim_2 | fact2
I want to merge them to have something like the following:
time_dim | fact1 | fact2
What I managed to get so far is something like this:
time_dim1 | time_dim2 | fact1 | fact2 (with, if we suppose there are 12 rows in time_dim1 and time_dim2, 12*12 = 144 rows instead of just 12)
Here is some more explanations about how I set my repository:
My fact table "issues" is made of the following columns:
issue_id, issue_type, issue_create_date, issue_end_date
I also have a table "calendar" with year, quarter, month and full_gregorian_date.
In the physical layer I created 2 aliases for the "calendar" table to join with the fact table using "issue_create_date" and "issue_end_date".
In the business model I created the 2 related time dimensions.
Everything is working fine so far, I was able to get 2 different analyses :
- time dimension (joined with "issue_CREATE_date") | count of issues (with aggregation rule "count" on issue_id)
to see how many new issues are CREATED through time (are there more issues created or less)
- time dimension (joined with "issue_END_date") | count of issues (with aggregation rule "count" on issue_id)
to see how many new issues are ENDED through time (when are more issues ended)
What I want is a third analysis like this:
- time dimension | count of created issues | count for ended issues
Thanks for your help.
Florence

did you set the content level as well?

Parent-child dimensions limitations

Hi eb
It is very convinient to develop complex heirarchized dimensions as parent-child especialy when they retain historical
changes in members.
I know one reason to not using such dimensions when they contain some 200,000 members or more due to performance issues (no aggregations on parent attributes???).
Are there any other reasons to not using parent-child heirarchies in ssas MD cubes???
TIA!
Rea

Hi reapeleg,
According to your description, you want to know weakness of using parent-child dimension. Right?
In Analysis Services, a parent-child dimension is distinguished by fact that it contains a hierarchy based on a recursive relationship. So it can only cause some performance problem. But the main reason of using a parent-child dimension
or not depends on the relationship between the attributes in your dimension. And there's workaround for the performance issue. Please see:
Parent-Child Dimensions in Analysis Services – Performance Walkthrough
Reference:
Parent-Child Dimensions
If you have any question, please feel free to ask.
Simon Hou
TechNet Community Support

Synchronizing data from two folders

It would be nice to be able to synchronize the description and keyword fields(possibly others) from two folders containing the same images(file numbers, different extension).
I keyword my images right after importing in Bridge and then convert them to .tif. Every now and then I add keywords to some .tif files that I feel needs them. This makes my .tif files different than my RAW files (either .dng or .psd). Having a synchronize window where you could select the folder with the updated keywords and then the one that you want to update and simply "hit" synchronize would be very handy.

Hi Leah
You could be in a fan trap situation. Is this possible? Could you perhaps have disabled the fan trap protection? If you do this and there is a fan trap then you can have issues with aggregations not working. This is correct too. I have known companies where they then export to Excel, get Excel to add the values and then say this is proof that Discoverer is no good. I tell them to go and look at the values manually and usually find that the answer in Excel is wrong.
If you are not in a fan trap situation you need to look at the underlying objects. I know you said you have an outer join which is fine. Now outer joins will bring back NULL values when there are no entries in the joined to object. If you do have such values being returned then you may not be able to aggregate them because if you try to SUM values with a NULL the answer will be NULL.
One way to knock out NULL values in an OUTER joined object is to use a Boolean OR+ condition. Let's say you have two folders, A and B, and you want to exclude values from B where say the AMOUNT is NULL. This condition will do the trick:
B.AMOUNT IS NOT NULL OR B.AMOUNT IS NOT NULL
I know it looks crazy but Oracle wrote special code into Discoverer that looks for this situation. Discoverer now rewrites the query to use a subquery thus overcoming the OUTER join restriction.
By the way this works too: B.AMOUNT IS NULL OR B.AMOUNT IS NULL - this one only brings back rows which have NULL in B.
Best wishes
Michael

Combining AS2 animation in AS3 CP4 project

I realize this is a no-no, and the message told me so in CP4, but I was able to get an AS2 built flash animation to work and publish in an AS3 CP project. The message, if I remember right, was 'you may experience unexpected results' or something to that effect. My question is: if I can get them to work as animated intros to the modules running as standalones, why can't I get them to work in an Aggreagtor file? What happens when aggreagted is that the flash intros simple bog down. They start out playing correctly the first time, but as you navigate the TOC the animated intro starts playing slower and slower. It's as if the previous version are still playing the in the background and just run out of memory/cache or something.
What I'm wondering is, could the problem be that the animations just simply aren't stopping? If so, is it possible to insert a command to stop?
Anyone have any ideas?
JT

I was able to get them to play 'satisfactorily' in CP 4 in single .swf- Not aggreagated. The larger problem was in Aggreagator. I had huge issues with aggregator, and adobe told me to upgrade to CP5 to solve that problem. Not sure that probelm is solved because now I have even more problems with the existing files-captions, graphics not displaying correctly. So for the time being, this AS2 AS3 issue is on the sidelines.
If I ever get to that oint again, I'll see If I can send a link.
Thanks!
JT

WebI issue with hierarchy display and aggregation

Trying to wrangle what looks like a defect in WebI's handling of hierarchy display and aggregation. We just completed an update cycle and are running BOBJ 4.1 SP4.
The hierarchy is a standard FM Commitment Item hierarchy in which both the nodes and leaves are Commitment Items (i.e. it uses InfoObject nodes, not text nodes). An example of one of these nodes looks like this in BW:
Cmmt_Item A - Node
    Cmmt_Item B - Leaf
    Cmmt_Item C - Leaf
Let's pretend Commitment Item A has $50 posted to it, B has $20 and C has $30. Analysis for OLAP handles this by adding a virtual leaf line to distinguish postings that are on the parent node like so:
Cmmt_Item A - Node       $100
    Cmmt_Item A - Leaf    $50
    Cmmt_Item B - Leaf    $20
    Cmmt_Item C - Leaf    $30
So you see both the total for the node ($100) and a line for each Commitment Items with KFs posted to them. Our users like this. They can easily see the aggregation and the breakdown.
WebI, on the other hand, will display it like this:
Cmmt_Item A - Node       $150
    Cmmt_Item B - Leaf    $20
    Cmmt_Item C - Leaf    $30
It doesn't create a separate line for the value of the parent node, but it does add it's value into the aggregate. Twice. Modifying the table with the 'avoid duplicate row aggregation' checkbox yields output like this:
Cmmt_Item A - Node       $100
Cmmt_Item A - Node        $50
    Cmmt_Item B - Leaf    $20
    Cmmt_Item C - Leaf    $30
We're about halfway there. While the top row now shows the correct aggregation and it creates a new line to show the distinct amount on the parent node, that new line appears on the same level as the parent. It's no longer clear that there's an aggregate and a breakdown. And attempting to expand or contract a node will now crash the report with one of those 'Error 16' messages.
Has anyone encountered this issue with hierarchies in WebI? This report was built from scratch in 4.1, so I'm not sure if this affects older versions or not. Or if it would affect any hierarchy that uses InfoObject nodes instead of text nodes.

Without a fix, the simplest workaround I can think of would be to restructure the hierarchy. It can't use postable nodes, so Cmmt_Item A - Node from my example would need to be converted into a text node and the postable characteristic added as a child on the same level as the B and C leaves.
This looks like it would affect anyone using hierarchies with postable nodes in a WebI report.
Another oddity in WebI's behavior here - even though the postable nodes show incorrect sums the sum at the root node is correct. So extending my examples from the original post:
Root Node                    $100
    Cmmt_Item A - Node       $150
        Cmmt_Item B - Leaf    $20
        Cmmt_Item C - Leaf    $30

Issues in aggregation

Similar Messages

Maybe you are looking for