Comments for Paul te Braak

Comment on SSAS Relationships and Granularity by Pearl

Pearl — Thu, 02 May 2013 21:20:00 +0000

Pretty nice post. I just stumbled upon your blog and wanted to
say that I’ve really enjoyed surfing around your blog posts. After all I’ll be subscribing
to your rss feed and I hope you write again soon!

Comment on DAX III – Filtering Queries with CALCULATETABLE and FILTER by buy Edu links

buy Edu links — Thu, 02 May 2013 17:57:55 +0000

Its not my first time to visit this web site, i am browsing this web
page dailly and get fastidious data from here every day.

Comment on Creating Custom Calculations in Excel with Pivot Table Extensions by Siraj Samsudeen

Siraj Samsudeen — Thu, 18 Apr 2013 13:43:56 +0000

Excellent article on how we can use this extension even on simple pivot tables. The tip about extracting the fully qualified name from GETPIVOTDATA is wonderful. Thanks for sharing.

Comment on Managing Connections In Power Pivot by Jim `Hutch` Hutcheson

Jim `Hutch` Hutcheson — Fri, 12 Apr 2013 18:20:06 +0000

Very strange, have to have version 2 of Powerpivot for Excel 2010 add-in installed and everything works fine, If version different – issues, no add-in issues. Any good write-up talking about restrictions between versions?

Comment on Managing Connections In Power Pivot by Paul te Braak

Paul te Braak — Fri, 12 Apr 2013 18:07:01 +0000

Can you open the PPVT widow and see the data model?

Comment on Managing Connections In Power Pivot by Jim Hutcheson

Jim Hutcheson — Fri, 12 Apr 2013 18:02:07 +0000

I cut/pasted the data into a blank excel sheet, re-did me slicers, etc. I made sure there were no connections listed. When I open the spreadsheet up in Excel 2010 with and without powerpivot add-in get error stating Initialization of the data source failed.

Comment on Designing Good Dimensions in SSAS by Paul te Braak

Paul te Braak — Fri, 12 Apr 2013 17:03:40 +0000

Hi Bhavik,

Devils always in the detail! – But I think this is a classic type 6 dimension – no? So retain the type 2 data for the hierarchy (historic) and type 1 for the ‘current [level 3] org hierarchy’ (attribute hierarchy) based on the name.

In your report, you could always filter by the hierarchy level and list the user attribute members that exist (which would ensure showing a single name).

However (as you nodoubt know), this (in my opinion) is really a training exercise for adhoc users. They need to be made aware that the two sets of attributes are not the same.

HTH, Paul

Comment on Managing Connections In Power Pivot by Paul te Braak

Paul te Braak — Fri, 12 Apr 2013 16:47:13 +0000

Hi Jim,

Not in Excel, although if you are sure that the data will not change, then perhaps just paste it in (this will not create a connection). If you are going to deploy the model to the server as an SSAS DB (with linked s/s), you’ll lose the connection anyway.

HTH, Paul

Comment on Managing Connections In Power Pivot by Jim Hutcheson

Jim Hutcheson — Fri, 12 Apr 2013 16:39:06 +0000

Is there a way to delete the connections after the initial load of data, so you load from MS Access – don’t need to refresh ever again. Doesn’t seem to let me delete the connection. I thought info stored in worksheet, just don’t want spreadsheet to ever try to link to connection again.

Comment on Designing Good Dimensions in SSAS by bhavikmerchant

bhavikmerchant — Thu, 11 Apr 2013 14:10:11 +0000

Hi Paul,

Havent seen you in a while! Hope Brissie is treating you well

I have an issue im currently working on with a similar scenario. We have an SCD2 dimension where there are duplicate names caused by current version and older versions of a business unit which are part of a 3 level org hierarchy. The older versions belong to a different parent. The key is composite on 4 attributes.. the 3 “level” keys and an “is_current” flag.

We need results across the version to show up on a single row (since the name is still the same), but not just for a chosen member. The query needs to return the whole org hierarchy, flattened to not have duplicates.

Furthermore this issue is complicated by the fact that this is in an SSRS report and the users can select any items at any level from the org hierarchy as a parameter to the report. I currently have the parameter dataset restricted to just the current versions so as not to show the users duplicates.

The problem I am have is how to get the main report query to accept those selections (in a STRTOSET), but return all the relevant data with no duplicates showing up.

I haven’t spent much time on it but hit a few roadblocks and am tackling it again next week.
Can you think of anything?