Tabular SSAS Models in Profitbase Studio

! The tabular model in PS6 was changed from build no. 5573 / 24.SEP.2015. Theese document is based on newer version.

Tabular Models

Tabular models are in-memory databases in Microsoft Analysis Services (SSAS), and is also available in Azure.

It is a columnar database capable of high performance and compression ratio. Different fact and dimension tables can be set up in a star/snowflake model so that the clients (e.g. Microsoft Excel and Microsoft PowerBI) can access them fast. Data Analysis eXpressions (DAX) is used for formulas and queries which is similar to using excel formulas.

Analysis Services tabular models support most of the same query languages that are provided for access to multidimensional models (MDX). The exception is tabular models that have been deployed in DirectQuery mode, which do not retrieve data from an Analysis Services data store, but retrieve data directly from a SQL Server data source. You can not query these models using MDX, but must use a client that supports conversion of DAX expressions to Transact-SQL statements.

It is not a good idea to use the tabular model when sources are based on dimensional modeling and has complex relationships with very large volume of data, or when you want complex calculations, scoping and named sets.

Standard Profitbase dimensions as Timefunction and Report is not supported directly. Parent/Child dimensions with default members etc. has limited usage.

With SQL 2019 (and forward), Calculation Groups addresses issues with time intelligence, currency conversions and the exploding number of meassures in a model in general. When using time intelligence, you e.g. gets Ammount, Ammount YTD, Ammount MTD, Ammount LY etc. For the user, this is overwhelming, and it becomes difficult to maintain both model and DAX expresions.
With Calculation Groups, reusable Calculation Items can be applied on any measure. 
This is especially useful for time calculations, where you can add a group with YTD, LY etc.
In Power BI, you can drag this group onto a measure (e.g. Ammount) and then get a set of calculated columns (Ammount YTD, Ammount LY etc.).

Tutorials