.NET Tutorials, Forums, Interview Questions And Answers
Welcome :Guest
Sign In
Win Surprise Gifts!!!

Top 5 Contributors of the Month
david stephan
Gaurav Pal
Post New Web Links

Cube Writeback with weighted allocation working strangely - creating large negative values against f

Posted By:      Posted Date: December 04, 2010    Points: 0   Category :Sql Server



I've got a cube (2008) with 2 MGs: Actuals and Budgets.

The Budgets partition is Write-Enabled.


On Excel 2010 the user will select the Actual Date for 2010-07, and will set the Budget date to 2011.

After seeing what the YTD actual figure looks like he enters the budget for 2011.

It is supposed to take that budget figure and split it between the funds that have actuals against them (using weighting).


Here's the problem:


SSAS allocates inflated figures against those funds, and then, so that the sum of those values will equal the originally entered amount, it puts a large negative value against the first Member in the fund dimension (I experimented with this by deleting the first member in that dimension. It just allocates the negative value against the new member!).

Why isn't it just allocating the values based on the ratio that it gets from that Actuals calc?



The UPDATE statement on the trace looked like this:


UPDATE CUBE [Forecasting]
SET  (
        [Actuals Date].[Year-Quarter-Month-Day Hierarchy].[Calendar Year Month].&[2010-07]
        , [Budget Date].[Year-Quarter-Month-Day Hierarchy].[Calendar Year].&[2011]

View Complete Post

More Related Resource Links

Writeback expression question - Weighted Allocation (SSAS 2008 SP2)




I'm trying to nail down the behaviour of writeback.

The Expression needs to change depending on how the value should be allocated.


E.g. If I want to spread a value to the leaf level funds then:



[Fund.Currentmember, Measures.Value] / [Fund.All, Measures.Value]


I've noticed that the expression should cater for each dimension that needs to be spread.

So if it is by fund, and by product then it should have:


[Fund.Currentmember, Product.CurrentMember, Measures.Value]


[Fund.All, Product.All, Measures.Value]


I've also found that if I need to allocate a value to an actual leaf level value, then the denominator should also carry a current member for that dimension. (I don't want to divide the value of the current member by the all level)


Eg. If I want to capture a value against a particular fund, then the weighted expression would look like this:


[Fund.Currentmember, Product.CurrentMember, Measures.Value]


[Fund.Currentmember, Product.All, Measures.Value]


What I'm trying to figure out is this:


What if I want to spread a value by a grouping attribute?

Example, if there was a Fund Group attribute that held multipl

Dynamically creating Gridview Template - Linkbutton Click not working



I have written some code to dynamically generate template columns for gridview which works well. However, each cell in the gridview has to be a linkbutton, which when clicked does a db update and redirects to a specific url with some parameters in it.

I have attached a click event handler to the linkbutton in the InstantiateIn method but the event does not seem to fire.

Could someone please help? the code is below -

//Dynamically creating the Grid

protected void btnAnalyze_Click(object sender, EventArgs e)

Creating a 3D cube and display content on each side

Dear all, I would like to show different content in each of a cube face that I can manipulated in space. My customer has see this in a desktop 3 D application somewhere on the net and would like to have this integrated in my applciation I am really interresting to integrate this has it is a good customer but have never done it Does any one have a done this already or sample for doing it ? regards SergeYour knowledge is enhanced by that of others.

Cube Writeback SQL Server 2008 R2 (SSAS, write back)

CUBE WRITE BACK in SQL SERVER 2008 R2 Did anybody get the SSAS write back functionality to work against a decent sized datawarehouse? I'm not asking about a little demo but a significant sized footprint e.g. 14M Rows measures, 6-8 wired Dimensions with 8-10 attributes each. I setup a test server using a Dell XEON ( 2x4core) with 48 GB RAM hardware and the latest Sql Server 2008 R2 release. I used the What-If scenario in Excel 2010 and modified a higher level, hoping SSAS would push the values down. After about 15 minutes and after having used 48 GB memory (on average 5% CPU) Excel throw an error - short on memory - and stopped the action. I found the same behavior in Sql server 2005 and 2008 and was hoping this would work now but apparently not so. Here are my questions: Is anybody using this at all? how can one calculated (roughtly) how much memory is requiered? is there any paper that describes best practise? Thanks for any help in advance, Dirk  

Negative values sums up incorrectly

Hello, I have calculated measure that removes negative values. Problem is the Total is summing up incorrectly. Here the script of the measure. Create Member currentcube.[Measure].[Amount2] as iif ([Measure].[Amount]>0, [Measure].[Amount], null)   Below is sample result. Customer        Amount      Amount2 A                    100            100 B                    -50             C                    100            100 D                    100            100 E                     100            100 TOTAL          &n

query negative and positive values and not their boolean counterpart

Good Day Need help.. I have this simple table, CREATE TABLE [dbo].[sample] ( [id] [int] NOT NULL , [number] [int] NULL , ) ON [PRIMARY] i wanted to make a select query that would display two columns(positive, negative) the sql statement that i made: "select (sample.number > 0) as positive, (sample.number < 0) as negative from sample " what happened is that, it displays the boolean counterpart and not the value.. I wanted to show this kind of data: substitute the "False" value into 0 and "-1/true" value with the exact absolute numerical value.   |debit||credit| |   0   ||  200 | |   0   ||  57   | |   50 ||  0     |   is this possible?  

Cube is showing empty values.

Hello,          When I do the full process for the cube, the data present the fact table getting updated properly into the cube and I had done the full process on the dimension which are part of the cube. But, when I tried to browse the cube, I am getting the empty values. Please help me on this. Thank you!! Regards, Pradeep.  

Questions about creating a class that returns multiple values


Hey all,

I am a little stumped, been trying to find some references to this but I am coming up with more confusion than anything else.lol

Here is my scenario using asp.net and VB.

I have an aspx page with about 10 label controls on it and a couple of image controls. Their data will be all populated dynamically by a single URL query string that is passed to a 3rd party url query string that retrieves an xml document with all of the information on it. I am currently just using the page load event and from that I can get everything I need and populate the page elements no problem with all of the code existing on the aspx.vb page.

However I thought for this project as I go forward I would like to create a vb class that handles all of the work instead of using the code behind page which has become quit convoluted.

I am not completely new to creating classes, I've done them before, and some pretty complex however this one is new due to the fact that I want to return multiple values.

So my thought is to approach it this way.

1.       On the page load event I grab the URL query string, and pass it to the vb class to begin the process.

2.       In the class I create a function that loads the xml document into memory.

3.     &n

Creating a Form say grid sort of a interface where in data values will be extracted from the table 4


how dis can be done help please with simple button click and changing the values it can be done but how to move columns with data in graceful manner the way we have in calendar control... 

Creating Local cube on Cube using Many-to-Many dimension relationship


Hi All,

I have got a cube that uses many-to-many dimensions relationship. I also have a calculated measure (A*B) in the measure group using a unary operator.

I tried creating the local cube by using CREATE GLOBAL CUBE statement and included the intermediate measure group and one measure from the group. I have also included all the dimensions that are linked to the intermediate measure group.

The local cube is being created but I don't see the correct values of the measures. I just get the value of A coming for the value of the calculated measures. There are also some performance issues while browsing the local cube which I assumed should be very fast.

I am not sure if this is a known issue and there's a solution available for this but I tried and didn't find anything on the forum. 

Please help me with this. Thanks in Advance.


Creating Local cube file using ASSL


Hi All,

I am looking to create a local cube by using ASSL. I didn't find much help on this. Please brief me the steps on how to create the local cube file using ASSL.

Thanks in Advance.

guide lines for implementing very large cube


Hey eb

I am in the process of designing a daily partitioned cube where each

partition should hold some half a milion records.

Cube also contain some 12 dimensions and that number may grow.

Are you aware of any guidelines i should follow to ensure cube is

stable and efficient?

I am using 32bit 2003 sp2 server with 32 GB RAM.

SSAS 2005 sp3.

I have already trouble processing (incrementaly) another smaller daily cube.

How can i allow ssas consume more RAM then its default settings?

As for quries perforemance are there any guidelines on top of

fine tuning aggregations usage and upgrading server to 64bit?

Thanks alot!


Publishin page layout, that add/update values in a metadata of the page not working


We have custom control on Publishin page layout, that add/update values in a metadata of the page. Code is written in 'Button click' event that update/add new values to the metadata. Following is the code snippet, which is working fine in Stage and Acceptance servers. Publishin page doesn't update Draft version with SPListItem.SystemUpdate instead it updates to last published major version.

currentPublishingPage = PublishingPage.GetPubl

creating SSAS cube and dimentions from OLTP

I have an interesting case. I need to create a dw solution for an OLTP schema. Is it possible to create a dw solution using SSAS while being sourced using an OLTP? Thanks.

Update with set of values not working

I am trying to do something like this http://weblogs.sqlteam.com/tarad/archive/2004/09/14/2077.aspx where you update a set of records with another set of records within a trigger.
I think the below update statement is close, but not correct.



tc SET LibraryChoice_RecordI

Are there any limitations in SharePoint 2010 BCS when working with large data?



We are working on suggesting the best approach that would be suitable to support user data stored in SQL Server 2008. User data is huge (in Terabytes). I need to understand if there are any issues using MOSS 2007 BDC or SharePoint 2010 BCS to interact with SQL Server 2008 database tables? I understand that BCS offers a no-code solution for interacting with SQL tables, however when handling such large data, can we have better performance if Stored Procedures and ADO.Net is used to fetch and update the data?

All in all I need to understand the pros and cons in both these approaches. Any pointers, specifically to related real world case studies shall highly be appreciated. Thanks much!

ASP.NetWindows Application  .NET Framework  C#  VB.Net  ADO.Net  
Sql Server  SharePoint  Silverlight  Others  All   

Hall of Fame    Twitter   Terms of Service    Privacy Policy    Contact Us    Archives   Tell A Friend