Please, check out http://www.RapidOlap.com. As the name states it's an OLAP creation/automation product that allows to create OLAP in minutes, using best practices templates. The package consists of RapidOlap Automation Server and RapidOlap Builder. It fills the missing middleware/backend gap.
In practice, OLAP functions allow application developers to compose analytic business queries more easily and more efficiently. For example, moving averages and moving sums can be calculated over various intervals; aggregations and ranks can be reset as selected column values change; and complex ratios can be expressed in simple terms. I've gained a profitable knowledge from an e-book. The book is available on Amazon & you can buy an e-book directly from the publisher by writing OLAPWorldPress@gmail.com.
Beware when you decide to use OLAP instead of RDMBS while you want to use Tableau. Many beautiful and powerful Tableau features will loss, such as rename/alias the dimension member, grouping dimension member, binning, add trend lines, add relative date filter easily, use user filter on tableau server.
So, is the recommendation NOT to use an OLAP data source like SSAS?
If you follow the recent recommendation not to use OLAP for Tableau's back-end to it's logical extension, then you also wouldn't want to use any RDBMS, but rather just a spreadsheet, against which Tableau would have virtually limitless manipulation flexibility. Of course, I'm intentionally overstating my case and, as an avoid OLAP dev, I'm far from un-biassed.
Having said that, with OLAP, which many consider an optimal data source for visual analyses, you really have two platform-type choices:
* Full-blown OLAP (Examples: MS SQL Server Analysis Services OLAP, Oracle Essbase, et al): Exceptionally good data sources for Tableau, yet requiring very specific skills to build and best accomplished within the context of a DW/BI framework, often with multiple technical contributors.
* OLAP Rapid-Dev Substitutes: (Qlikview, MS PowerPivot, et al). Here, data-savvy analysts, without much dev assistance, can themselves rapidly build OLAP-like data structures against which Tableau can then provide killer vizzes. Upside: Much less formal dimensional modelling, data integration and data cleansing. Downside: See 'Upside'. Many view these substitutes as best suited to Departmental (vs. Enterprise-level) solutions.
As a note, I hear that Tableau now connects to PowerPivot.
I can confirm that Tableau connects to PowerPivot. With Tableau 5.2 you can connect to PowerPivot saved to SharePoint with the Excel Services and PowerPivot extensions. At the Tableau Customer Conference in August I demoed 6.0 and connected to a local PowerPivot .xlsx files -- you'll will need to install an Excel add-in which acts as Tableau's "driver" to PowerPivot. PowerPivot's speed is a nice match to for Tableau's interactivity.
I've noticed that if I enter values into the optional Display Folder property of cube measures in SSAS 2008, the Measure Groups collection no longer shows up in the measures area of Tableau. Instead, the Display Folders are displayed. It seems to override the Measure Groups. If so, this is an undesired behavior. The measure groups are usually the parent collection and the display folders, if enabled, are nested within it. Anyone in Tableau finding this behavior?
As a comparison, connect to a cube using Excel and it preserves Measure Group parent collection in the field list display. Any display folder gets nested.
Hi Julius, I also have the same behaviour. When Tableau connect to SSAS, it will display all measures in the connected cube and not group by display folder or measure group.
Have anyone here successfully connect Tableau v6 to PowerPivot xlsx? I always got the following error (don't know why):
[Tableau PowerPivot AddIn] Unable to connect to the MS PowerPivot data source. Couldn't find valid MS PowerPivot data in the workbook you specified.
Unable to connect to the server. Check that the server is running and that you have access privileges to the requested database.