DAX - Aggregating measures by multiple dimensions.

I have a pretty simple data warehouse (star schema). DimDate, DimProduct, DimCustomer, FactSalesOrders. I have created a Tabular Model with this data and have several measures in FactSalesOrders. Ultimately, I am trying to consume this data in SSRS
with a DAX query. Therefore, I need to perform a "group by" function in DAX to group by several attributes from each of the Dimensions and aggregate the measures in the FactSalesOrders table. I've tried summarize, and addcolumns, but the performance
is slow. 10 seconds to pull a simple "group by" DimDate.FiscalYear, DimProduct.SKU, DimCustomer.CustomerName. Analysis Services within Excel is able to slice this data in less than a second. I'm looking for some insight as to how to write the
DAX query.

Hi Mstaples123542,
According to your description, you want to create "group by" function in DAX query. Right?
In DAX, the only way to achieve "group by" is using summarize, addcolumns. There's no better way the get the same result with better performance. Since you are using SSRS, I suggest you just use the DAX to get the whole dataset without any group
and order, then you group and order all data in the SSRS.
If you have any question, please feel free to ask.
Regards,
Simon Hou
TechNet Community Support

Similar Messages

AWM Set Aggregation Operator based on Measure instead of Dimension?

Is there anyway to base an operator of a measure over a dimension on the measure instead of the dimension? In AWM, I notice that each Measure has an Aggregation tab that summarizes its operator over each dimension, but it is greyed out. The way I'd like it to work is to have one Measure be summed over the Dimensions, while another Measure is averaged over the Dimensions.

Found my answer: Not only do I need to make sure my cube is not compressed, but I also need to select "Override the Aggregation specification of the cube" under each Measure's General tab. This results in the contents of the Aggregation tab no longer being grayed out.

AWM Calculated Measure for Percent of Total Across Multiple Dimensions?

I noticed that AWM has a Share function that gives me a percent total of a grain from a specific hierarchy's Top of Hierarchy, but is there any way to do that with multiple dimensions? For example, if I had a Share of Dimension X = 55% Where Time = 1/1/2013 and a Share of Dimension Y = 32% Where Time = 1/1/2013, then could I have a Share of both Dimension X and Dimension Y = 16% Where Time = 1/1/2013?

I had used an alternate solution which is a bit more cumbersome using native OLAP_DML formulae/expression.
The above expression using OLAP Expression Syntax is much better (if it works).
I was not aware of this OLAP expression syntax when i needed to create a kpi for similar requirement.
We had Qty measure and Share along dimension like "Qty - Share at Year level", "Qty - Share of Total Customer".
We needed to get "Qty - Share at Total Customer, Year".
Cube= SALESCUBE
Base Meas = QTY
Step 1) Create Measure which represents "Qty - Total Customer, Year" which will work in any reporting context ...
Note: For time=day1/2/3/ within same year and customer=cust1/2/3/... or Total Customer, the expression result will be constant (result fixed for any dimension members/status along the 2 dimensions - TIME and CUST). Denominator Value changes each year since we have constrainted time dimension to the ancestor of current cell at YR level. If we choose anscestor at TOP level TOTTIME say then the value is fixed for all time dimension members/values.
olap dml expression: QUAL(SALESCUBE_QTY, CUSTOMER limit(CUSTOMER to CUSTOMER_LEVELREL 'TCUST'), TIME limit(limit(TIME to ANCESTORS USING TIME_PARENTREL TIME(TIME TIME)) KEEP TIME_LEVELREL eq 'YR'))
in awxml - this becomes:
ETMeasureColumnName="QTY_TCUST_YR"
Name="QTY_TCUST_YR"
MeasureExpression="OLAP_DML_EXPRESSION('QUAL(SALESCUBE_QTY, CUSTOMER limit(CUSTOMER to CUSTOMER_LEVELREL ''TCUST''), TIME limit(limit(TIME to ANCESTORS USING TIME_PARENTREL TIME(TIME TIME)) KEEP TIME_LEVELREL eq ''YR''))', NUMBER)">
<Classification
Value="AwmDescriptionType=OLAP_DML_CALC"/>
<Description
Type="LongDescription"
Language="AMERICAN"
Value="Qty - Total Customer, Year">
Step 2) Define the share measure explicitly since we have already calculated the denominator needed for composite share.
Check for division by 0 error before performing the share calculation explicitly as Numerator= Qty (for current cell/reporting context) and Denominator = Qty - Total Customer, Year (from Step 1)
olap dml expression: if SALESCUBE_QTY_TCUST_YR ne 0 then SALESCUBE_QTY / SALESCUBE_QTY_TCUST_YR else na
in awxml - this becomes:
ETMeasureColumnName="QTY_SHARE_TCUST_YR"
Name="QTY_SHARE_TCUST_YR"
MeasureExpression="OLAP_DML_EXPRESSION('if SALESCUBE_QTY_TCUST_YR ne 0 then SALESCUBE_QTY / SALESCUBE_QTY_TCUST_YR else na', NUMBER)">
<Classification
Value="AwmDescriptionType=OLAP_DML_CALC"/>
<Description
Type="LongDescription"
Language="AMERICAN"
Value="Qty - Share of Total Customer, Year">
</Description>
Report needs to use measure QTY_SHARE_TCUST_YR (Qty - Share of Total Customer, Year) defined in Step 2. It may be useful to expose/display intermediate measure QTY_TCUST_YR also so as to make the basis of calculation very clear to the user.
Nick,
If you customize above soln. to use Total Time, Total Prod and Total Organization (along 3 dimensions)... I am sure the fact table has a time dimension which should also be factored in in your calc/defn.
If you have missed out the join to Time dimension from fact in your queries, in olap reporting terms, its similar to Time Dimension selection of Time level = TTIME Total Time (1 node at TOP).
You need:
Step 1) Qty - Total Time, Total Product and Total Org
Step 2) Qty - Share of Total Time, Total Product and Total Org
Then the example you gave should be covered via QTY and QTY_SHARE_TTIME_TPROD_TORG (Step 2)
SUM(quantity)
2875should be QTY, QTY_SHARE_TTIME_TPROD_TORG (=100%) at Time level = TTIME, Product level = TPROD, Org level = TORG
SUM(quantity)
345 [12% of all products sold]should be QTY, QTY_SHARE_TTIME_TPROD_TORG (=12% hopefully) at Time level = TTIME, Product level = PROD with report filter on product = 'CX-867054', Org level = TORG
SUM(quantity)
977 [34% of all products sold]should be QTY, QTY_SHARE_TTIME_TPROD_TORG (=34% hopefully) at Time level = TTIME, Product level = TPROD, Org level = STORE with report filter on store = 'NY_ALBA_013'
SUM(quantity)
88 [3.06% of all products sold]should be QTY, QTY_SHARE_TTIME_TPROD_TORG (=3.06% hopefully) at Time level = TTIME, Product level = PROD with report filter on product = 'CX-867054', Org level = STORE with report filter on store = 'NY_ALBA_013'
HTH
Shankar

How to define an aggregation rule for a dimension based on bridge table?

Hello,
I need a solution for aggregating data correctly when using a dimension based on a set of dimensione tables containing a bridge table. Please find below a description of my business case and the OBIEE model which I’ve created thus far.
Business Case
The company involved wants to report on the number of support cases, the different types of actions that were taken and the people involved in those actions. One support case will undergo a number of actions (called ‘handelingen’) until it is closed. For each action at least one person is involved performing a specific role, but there can also be multiple persons involved with 1 action, each performing a different role for that action. This is the N : N part of the model.
The problem that I face is visible in the two pictures below:
http://i84.photobucket.com/albums/k24/The_Dutchman_2006/OBIEE/sample.png
As long as I don’t include anything from the Dimension Meelezer in my report, I get the correct number of handelingen (7). When I include the person (called ‘Meelezer’), the measuere per action is multiplied by the number of persons/roles involved with that action.
When I changed the Aggregation rule in the report column #Handelingen to ‘Server Complex Aggregate’ I do get the correct endtotal:
http://i84.photobucket.com/albums/k24/The_Dutchman_2006/OBIEE/sample2.png
I believe it should be possible to define in the repository a different aggregation rule for individual dimensions, but I’ve not been able to achieve this.
Explained below is what I have created in my Physical and Business Model & Mapping layers:
The Physical Model is built like this:
(This is just a small part of a much larger physical model, but I’ve only included the most relevant tables)
http://i84.photobucket.com/albums/k24/The_Dutchman_2006/OBIEE/PhysicalDiagram-1.png
The Fact table (ALS Feit Zaakverloop) contains FK’s for the action (FK_HANDELING, joined to ALS Dim Handeling), the date the action took place (FK_DATUM_ZAAKVERLOOP, joined to ALS Dim Datum Zaakverloop) and the uniqe group of people involved (FK_MEELEZERS, joined to ALS Groep Meelezers) and a measure column (SUM_HANDELINGEN) populated with the value ‘1’ for each row.
The Bridge table (ALS Brug Meelezer/Reden Meelezen) contains three FK’s: FK_GR_MEELEZERS (joined to ALS Groep Meelezers), FK_MEELEZER (joined to ALS Dim Functionaris) and FK_REDEN_MEELEZEN (joined to ALS Dim Reden Meelezen).
The Business Model
In the business model, the four physical tables for the N:N relation have been combined into one logical dimension table.
http://i84.photobucket.com/albums/k24/The_Dutchman_2006/OBIEE/BusinessModel-1.png
DIM Meelezer contains one LTS in which the four physical tables have been combined:
http://i84.photobucket.com/albums/k24/The_Dutchman_2006/OBIEE/LTS1.png
And all the required locical columns have been created:
http://i84.photobucket.com/albums/k24/The_Dutchman_2006/OBIEE/LTS2.png
DIM Meelezer has also been identified as a bridge table and a Business Key has been defined on a combination of the FK’s in the bridge table and business codes of the two dimension tables.
http://i84.photobucket.com/albums/k24/The_Dutchman_2006/OBIEE/BMDIM.png
Next a hierachy was created for Dim Meelezer:
http://i84.photobucket.com/albums/k24/The_Dutchman_2006/OBIEE/Hier.png
In Feit Zaakverloop, a measurement called ‘# Handelingen’ was created using SUM_HANDELINGEN, with an aggregation rule of SUM.
In the LTS of both the DIM Meelezer and Feit Zaakverloop, the Logical Content Levels have both been set to: LVL Detail – Meelezer.
Please provide suggestions that will NOT require changes to the physical datamodel as they would require too much time to achieve (or at leats would not be ready before my deadline.
Thanks!
Edited by: The_Dutchman on Dec 13, 2011 11:43 AM

Hmm, no replies yet...
Am I in 'uncharted territory' with this issue?

OBIEE Error[14041] Nested Aggregation measure are currently not supported

Hi,
Please provide workaround for the OBIEE Error[14041] Error in measure definiion. Nested aggregation measures are currently not supported.
I have two logical columns - Current YTD Invoice Quantity and Prior YTD Invoice Price. I want to create a column which is -
(Current YTD Invoice Quantity * Prior YTD Invoice Price)/ sum(Current YTD Invoice Quantity * Prior YTD Invoice Price)
sum in the denominator should be the sum of all rows returned by the report. So level based measures cannot be used as there are multiple report and dimension used may vary.
Columns Current YTD Invoice Quantity and Prior YTD Invoice Price are already aggregated to Sum.
If I try to do the sum on the physical column while creating logical column, the query is running for 6-7 hours and not giving any result. Both the columns belong to different alias table in physical layer.
Please let me know if you guys know any work around for this issue. Also let me know whether this type of nested aggregation are supported in 11g or not?
Thanks.

Hi Adil,
Yes, I did create a hierarchy for the time dimesion.
No, I was not able to specify the aggregate rule since the source of the logical column is another logical column
But the time hierarchy works for rows that is not based upon the SUM aggregate at the answers level.
Say I have 3 columns: Income,Expense, Bottom Line (formula : Income- Expense)
Income is created based upon a case statement in the Logical Column and
Expense is also created based upon a case statement in the Logical Column.
Income 1000 Rupees
Expense -300 Rupees
Bottom Line 700 Rupees (where the Bottom line is a row based upon the Answers aggregate formula
which is SUM(Income+Expense))
When I add a column year, I get the following output
Income 2009 500 Rs
Income 2008 500 Rs
Expense 2009 -150 Rs
Expense 2008 -150 Rs
Bottom Line 2009 700 Rs
Bottom Line 2008 700 Rs
The Bottom Line doesn't spread across the Year
Hope this helps you to understand what my problems is.
Thank you!

Measure Compatibility with Dimensions

Hello Folks,
In Cognos, We have an option on defining measures compatibility with dimensions inside the cube i.e. incase if the cubes has measures from multiple fact tables along with the dimensions that are not common across fact tables, then the measure that are not compatible with dimensions can be defined, so that the measure value appears as NA when we probe the cube. Just wondering do we have anything euquivalent in Oracle OLAP.
Cheers...

Write iMovie Feedback
Let the iMovie teams hear what you have to say about iMovie 08.

Grouping of measures in accounts dimension Essbase

I am new to essbase .
I am developing a cube for one of the requirements which has got
following measures in accounts dimension
1) Allocated quantity
2) Picked Quantity
3) Shipped Quantity
4) Allocated Amount
5) Picked Amount
6) Shipped Amount
Now I want to group these measures in 2 groups viz a) Qtys b) Amounts
means
Qtys : Allocated quantity
Picked Quantity
Shipped Quantity
Amounts : Allocated Amount
Picked Amount
Shipped Amount
wanted to show these groups along with its measures in Excel add in report .
I am using Essbase Integration Services 11 to develop the metaoutline
How to define these groups in accounts dimension
Regards
Amit

If you have a way to differeniate these measures you can add user defined measures in EIS. One way you could differeniate them is to put in filtering sql the uses like clauses to select members with Quantity or amount in them and have them put under the headings you want

Report with multiple dimensions in Row and Column with EvEXP

Hi
We are currently running BPC 7.0 SP3 and try to create the report(or schedule) with more 2 x 2 dimensions in the row and column.
e.g.
row dimension : Outer row is Entity, Inner row is Account
column dimension : Outer column is Time, Inner row is Category
We created the report with 2 x 2 dimensions like above e.g. with EvEXP and EvNXP, NOT EvDRE
based on default dynamic template:Template13.xlt(Nested Row, 2 x 1 dimensions defined by EvEXP and EvNXP formula).
However, when expanding the dimension with double click these dimensions or change CV of the report with 2 x 2 dimensions defined by EvEXP and EvNXP formula,
error occurs on EvEXP and EvNXP formula and can't retrieve the result correctly.
[Question]
Can't we create the report with more 2 x 2 dimensions in the row and column using EvEXP and EvNXP function ?
We understand that we can make the report with more 2 x 2 dimensions in the row and column if using EvDRE.
Thank You in advance
Kenya

Thank you for your advice.
I understand that we should use EvDRE when creating the report with multiple dimension in the row/column　because EvEXP have some erros, poor performance, etc.
Then, we have two questions.
1)
>plus they had some errors in the designs.
what kind of errors they have as an example?
2)
>If you must build the EVEXP and EVNXP, I would suggest building the template from scratch
We built the simple template with just 2 x 2 dimension from scratch.
But this report doesn't work well when double-click the dimension or changing CV...
Would you please check the following template?
<https://sapmats-de.sap-ag.de/download/download.cgi?id=RCZE1ME4YSORY3DCDO4NNMANZLBL1L5ZBUOB2F71YVNVS8XDJW>
It would be greatly appreciated if you could give me your advice for the template.
Thank you again
And my best regards,
Kenya

Pie chart with two measures and date dimension navigation not working

Hi Experts,
Pie chart with two measures and date dimension navigation not working. Any help is appreciated.
Thanks
V

Hi Deepak,
I had time dimension in the RPD.
I have stacked bar chart with same time dim like year & month in the report. when I go to legand and set navigation it is working fine. But not with pie chart.
I am not not sure what is the problem. When I click on Pie chart it is not navigating to the target report. Can it be any other issues..???

Multiple dimensions from single logical table

Hi,
I cannot seem to create multiple dimensions based one one single logical table. For example, I have a Time table from which I'd like to create a calendar time dimension and a fiscal time dimension. However, as soon as one dimension is created, I cannot create another dimension from the same logical table.
Is it possible?
Thanks

Jlin,
If I understood your requirement correctly you need to create multiple Hierarchies out of a single Logical Dimension table, this is possibe the base rule for creating a hierarchy is you should have a common starting point( Grand Total Level) and a common ending point( Detail Level), first create this and create sublevels and share the Detail Level as a child. This will allow you to create multiple hiearchies out of a single dimension. One level will have drill down on the Calendar dimension and one one the Fiscal and both will have the common detail key like day key.Hope this helps.
Arjun

New calculated member based on existing measure and a dimension value

Hi
I’m trying to create a new calculated member in my cube based on an existing measure and a dimension value.
I can do it easily in DSW but I want to do it as a Calculated Member in Calculations tab from my cube.
I have a dimension called Product in which one attribute is UOM (unit of measure) and a measure called Quantity. The relationship with the fact table is “Regular”.
For each product we have a unit of measure and an associated quantity measure:
Product Name
UOM     Quantity
Product 1
          Km
         50
Product 2
          Kg
         60
Product 3
          Kw
          25
Product 4              L
15
Product 5              T
         1.5
Total                                     151.5
The calculated member I want to create is “Kw” and has to contain the value from Quantity measure when Product.UOM is “kW” and should look like this:
Product Name
UOM     Quantity
Kw
Product 1
          Km
         50
Product 2
          Kg
         60
Product 3
          Kw
          25
25
Product 4              L
             15
Product 5              T
         1.5
Total                                     151.5
25
So far I tried using the MDX with “IIF” or “Case” but with no results.
What statement I should use in this case?
Thank you.

Hi Cees,
Thank you for your answer.
I never used yet the “Scope” statement and yet it’s an elegant solution.
I use it as in your example:
CREATE
MEMBER CURRENTCUBE.[Measures].[Kw]
AS NULL,
FORMAT_STRING =
VISIBLE = 1 ,
ASSOCIATED_MEASURE_GROUP =
'TR Other';
Scope [Measures].[Kw];
Scope [Product].[UOM].[kW];
This = [Measures].[Quantity];
End
Scope;
End
Scope;
…and the calculations are solid…with one exception: when I took out UOM from the query there are no more any resuts:
Product Name    Quantity    Kw
Total
It’s only working
if using UOM in the query.
I also thought that a
tulpe might be another solution as:
([Measures].[Quantity],[Product].[UOM].&[kW])
On this last one I’ve got the expected results with one exception: when I use UOM in the query all the other UOM values are listed each one of them
having the [Kw] measure calculation value.
Product Name    UOM     Quantity    Kw
Product 1              Km
         50
Product 2              Kg
         60
Product 3              Km
          25            25
Product 3              Kg
           25            25
Product 3              Kw
          25            25
Product 3              L
25            25
Product 3              T
25
  25
Product 4              L
            15
Product 5              T
         1.5
Total                                     151.5
25
…and without UOM in the query I have what I’m expected:
Product Name    Quantity
Kw
Product 1              50
Product 2              60
Product 3               25
25
Product 4
15
Product 5
1.5
Total                       151.5
25
Maybe we can work together to find a solution.
Thank you,
Oks

Multiple Dimension Builds using MAXL

We currently use ESSCMND to perform multiple dimension builds and data loads. We have been using the "INCBUILDDIM" in Escmnd to update the outline with multiple files without restructuring until the last file. Is there an equivalent command for MAXL ??

Yes, you can mimic the BEGININCBUILDDIM command from ESSCMD in version 6.5. I've appended a sample script below to show that functionality. Until you upgrade to V6.5 you could shell to Esscmd and do your dim builds there. login 'xxx' 'PASSWORD' on 'LOCALHOST'; spool on to 'd:\ESSBASE\CLIENT\OUTPUT.LOG'; alter system load application 'SAMPTEST'; alter application 'SAMPTEST' load database 'BASIC'; import database 'SAMPTEST'.'BASIC' dimensions from server text data_file 'ACCTBLD1.TXT' using server rules_file 'ACTBLD.RUL', from server text data_file 'ACCTBLD2.TXT' using server rules_file 'ACTBLD.RUL' preserve all data on error append to 'ERR.OUT'; spool off; logout; exit;

Invalid navigation state. Multiple dimension values from model

We are getting following error on the endeca server in production environment and we are not able to reproduce it in the lower environment, can some body expains what this error means and how this occurs
WARN 03/17/13 20:59:39.418 UTC (1363553979418) DGRAPH {dgraph} Error processing HTTP exchange 797486: Invalid input : Invalid navigation state. Multiple dimension values from model (18,"Product Aisle") in query nav state.
WARN 03/17/13 21:00:27.575 UTC (1363554027575) DGRAPH {dgraph} Error processing HTTP exchange 798401: Invalid input : Invalid navigation state. Multiple dimension values from model (18,"Product Aisle") in query nav state.
WARN 03/17/13 21:01:31.394 UTC (1363554091394) DGRAPH {dgraph} Error processing HTTP exchange 800340: Invalid input : Invalid navigation state. Multiple dimension values from model (18,"Product Aisle") in query nav state.
WARN 03/17/13 21:01:43.216 UTC (1363554103216) DGRAPH {dgraph} Error processing HTTP exchange 800577: Invalid input : Invalid navigation state. Multiple dimension values from model (18,"Product Aisle") in query nav state.
WARN 03/17/13 21:02:39.912 UTC (1363554159912) DGRAPH {dgraph} Error processing HTTP exchange 801775: Invalid input : Invalid navigation state. Multiple dimension values from model (4,"Form") in query nav state.

Hello Tijomon,
Looking at the warning it seems the query being build is not good. Some possibilities are:
~ you have multi select enabled in the front end on a single select dimension
~ the index was moved from one environment to other and some of the dimension values do not match (due to the state being different)
HTH,
Pankaj.

Measurement for multiple numeric limit test

Hello,
I am developing a test software using LabWindows and TestStand. In my application, I have added a no: of multinumeric tests. In Operator Interface, I want to display each measurement in run time, instead of waiting for completion of all the measurements with Multiple numeric limit test. Is there any way to receive in TSOP each runtime measurent taken for multiple numeric limit test without waiting for multinumeric limit test over?
Thanks in advance......
Best Regards,
Vijay

Hi Vijay,
Unfortunately, a Multiple Numeric Limit Test will not allow you to return each measurement inside the step as it executes. If you need this to occur, there are other methods that you could take. You could create a custom step type or use many Numeric Limit Test steps and pass them the indexes in your array.
Another method you might use would be to use an Action step and save the array locally. Then, you could use Statement steps to evaluate each item in the array.
Jessica
National Instruments
Product Support Engineer

Is it possible to add multiples dimension objects in Webi Ranking ?

As per the project requirement we need to convert custom sqls to reporting sql in webi 3.1 sp5 .
Is it possible to add multiple dimension objects while applying ranking (Webi query panel ).
Or is there any other way to achieve this feature in BO ?
Regards,
Pranay

Hi ,
There is a way if you want to apply ranking within a ranking . Below are the steps:
First on the outer side we can use the Rank feature which is available directly
Then use the Rank Function (=rank ()) and reset your dimension accordingly .
This kind of Ranking is possible.
Regards,
Rahul

DAX - Aggregating measures by multiple dimensions.

Similar Messages

Maybe you are looking for