Tag: Tabular Model

Azure, BISM Tabular, Business Intelligence - BI, DAX, Excel, Power BI, Power BI Desktop, Power BI Embedded Capacity, Power BI Premium Capacity, Power BI Service, Self-Service BI, SQL Server, SQL Server Analysis Services - SSAS, SQL Server Management Studio - SSMS, Tableau Desktop

What Does XMLA Endpoints Mean for Power BI and How to Test it for Free?

Test Environment from Power BI XMLA Endpoint

XMLA endpoint connectivity for public preview has been announced late March 2019. As at today, it is only available for Power BI Premium capacity users. This sounds like a massive restriction to a lot of people who don’t have a Premium capacity, but they’d love to see how it works. In this article I show you an easy way to get your hands to Power BI XMLA endpoint as quick as possible. Before I start, I’d like to simply explain what XMLA endpoint is and what it really means for Power BI users.

Power BI is Like Onion! It has layers!

Generally speaking, Power BI has two different layers, presentation layer and data model layer. Presentation layer is the visual layer, the one you make all those compelling reports and visualisations. The data model as the name resembles, is the layer that you make your data model in. This layer is the one you can access it via XMLA connectivity.

In a Power BI Desktop file, you can see both layers:

Different layers of Power BI

How XMLA Relates to Different Layers in Power BI?

As you may have already guessed, XMLA is only related to the data model layer and it has nothing to do with the presentation layer. So you may connect to a data model, browse the data model, import data from the model to other platforms like Excel and so forth.

XMLA Is Not New!

Seriously? Yes, seriously. It is not new. It’s been around for many years and perhaps you’ve already used it zillions of times. Whenever you’re connecting to an instance of SQL Server Analysis Services, either Multidimensional or Tabular from any tools like SQL Server Management Studio (SSMS), Power BI Report Builder, Excel, Tableau, etc…, you’re using XMLA connectivity indeed.

Power BI is an Instance of SSAS Tabular

It is true. Power BI runs a local instance of SSAS Tabular model. So, whenever you open a Power BI Desktop file (PBIX), Power BI creates a local instance of SSAS Tabular model with a random local port number that can be accessed on your local machine only. When you close the file, the local instance of SSAS Tabular is shut down and its port number is released.

I first revealed the fact that you can connect to the underlying data model in Power BI Desktop from whole different range of tools like SSMS, SQL Server Profiler, Excel, etc… on Jun 2016. So, we indeed were using XMLA to connect to Power BI data models for a long time. We can even take a step further to import our Power BI data models into an instance of SSAS Tabular. In that sense, we are literally generating XMLA scripts from Power BI to create the same data model in SSAS Tabular. How cool is that?

Sooo… What is new then?

Continue reading “What Does XMLA Endpoints Mean for Power BI and How to Test it for Free?” →
DAX, Power BI, Power BI Desktop

Using “IN” Operator in DAX

IN operator in DAX

If you are a SQL guy I bet you’ve used “IN” operator zillions of times. You might also looked for the same functionality in DAX and I’m sure you’ve found fantastic blog posts showing you how to mimic the same functionality in DAX. The October release of Power BI Desktop is full of new analytics features such as Grouping, Binning and TOPN filtering. On top of that, one new awesome feature that is not documented at time of writing this article, or at least I haven’t find anything over the internet, is “IN” operator in DAX. In this post I show you how to use it in your DAX expressions.

Requirements

Note 1: You need to install SSMS2016 to be able to write DAX queries provided in this article. Alternatively, you can use DAX Studio . If for any reasons you cannot use SSMS 2016 or DAX Studio and you only have Power BI Desktop, don’t worry, I’ll provide some examples in Power BI Desktop as well.

Note 2: If you run previous versions of SQL Server it’s absolutely alright. There is nothing special in AdventureWorksDW2016CTP3 for this article that you don’t get in older versions of the sample database. But, keep in mind that SQL Server 2016 Developer Edition is now free and you can download it very easily. Check this out if you’re interested to see how.

Getting Started

After downloading the latest version of Power BI Desktop run it then

  • “Get Data” from SQL Server
  • From AdventureWorksDW2016CTP3 load “FactResellerSales”, “DimProduct”, “DimProductCategory”, “DimProductSubCategory” and “DimDate” to Power BI Desktop model
  • Find the local port of Power BI Desktop by opening “msmdsrv.port.txt” file from the following path:

“%UserProfile%\AppData\Local\Microsoft\Power BI Desktop\AnalysisServicesWorkspaces\AnalysisServicesWorkspaceXXXXXXXX\Data”

Note: The “XXXXXXXX” postfix is a random number. 

  • Open SSMS 2016 and connect to Power BI Desktop model as an Analysis Services local server. Do you want to learn more about how to connect your Power BI Desktop model from different software? Then check this out.

SSMS Connect to Power BI Desktop Model

  • Open an MDX new query
  • Run the following DAX query
EVALUATE
    SUMMARIZE('FactResellerSales'
                , DimDate[CalendarYear]
                , "Total Reseller Sales"
                , SUM('FactResellerSales'[SalesAmount])
                )

Here is the results:

Writing DAX in SSMS

Now we want to filter “CalendarYear” so that the query shows sales values for 2011 and 2012 only. One common scenario we had to do in prior versions of Power BI Desktop, Power Pivot or SSAS Tabular model was to use a logical OR operator “||” like below:

EVALUATE
FILTER(SUMMARIZE(FactResellerSales
                    , DimDate[CalendarYear]
                    , "Total Reseller Sales"
                    , sum(FactResellerSales[SalesAmount])
                    ) , DimDate[CalendarYear] = 2011 || DimDate[CalendarYear] = 2012
                    )

From now on we can write the above query using “IN” operator in DAX like below:

EVALUATE
    FILTER(
        SUMMARIZE(FactResellerSales
                    , DimDate[CalendarYear]
                    , "Total Reseller Sales"
                    , sum(FactResellerSales[SalesAmount])
                    ) 
            , DimDate[CalendarYear] 
                IN (2011, 2012)
            )

Here is the results:

IN operator in DAX

Continue reading “Using “IN” Operator in DAX” →