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?”

Quick Tips: How to Filter a Column by another Column from a Different Query in Power Query

Filter a Column by a Column from a Different Query in Power Query

A while ago I was visiting a customer that asked if they can filter a query data by a column from another query in Power BI. And I said of course you can. In this post I explain how that can be achieved in Power Query. The key point is to know how to reference a query and how to reference a column of that query in Power Query. This is useful when you have a lookup table that can be sourced from every supported data source in Power Query and you want to filter the results of another query by relevant column in the lookup query. In that case, you’ll have a sort of dynamic filtering. So, whenever you refresh your model if new records have been changed in or added to the source of the lookup query, your table will automatically include the new values in the filter step in Power Query.

Referencing a Query

It is quite simple, you just need to use the name of the query. If the query name contains special characters like space, then you need to wrap it with number sign and double quotes like #”QUERY_NAME”. So, if I want to reference another query, in a new blank query, then the Power Query (M) scripts would look like below:

let
    Source = Product
in
    Source

Or something like

let
    Source = #"Product Category"
in
    Source

Referencing a Column

Referencing a column is also quite simple. When you reference a column you need to mention the referencing query name, explained above, along with the column name in brackets. So, the format will look like #”QUERY_NAME”[COLUMN_NAME]. The result is a list of values of that particular column.

let
    Source = #"Product Category"[Product Category Name]
in
    Source
Referencing a Column from Another Query in Power Query
Continue reading “Quick Tips: How to Filter a Column by another Column from a Different Query in Power Query”

Quick Tips: Time Dimension with Granularity of Seconds in Power BI and Analysis Services Tabular Models

Quick Tips: Time Dimension with Granularity of Seconds in Power BI and Analysis Services Tabular Models
Photo by Markus Spiske

In an article I posted a while back I showed different methods of creating Time dimension in Power BI and Tabular models. The Time dimension I explained was in Minutes. In this post I show you simple way to create Time dimension supporting Seconds. As this is a quick tip, I only show you how to get the Time and ID columns in the Time dimension. If you need to add time bands (time buckets) check this out for more details.

Time Dimension in Seconds Grain with Power Query (M):

Copy/paste the code below in Advanced Editor to generate Time dimension in Power Query:

let
Source = Table.FromList({1..86400}, Splitter.SplitByNothing()),
#"Renamed Columns" = Table.RenameColumns(Source,{{"Column1", "ID"}}),
#"Time Column Added" = Table.AddColumn(#"Renamed Columns", "Time", each Time.From(#datetime(1970,1,1,0,0,0)+#duration(0,0,0,[ID]))),
#"Changed Type" = Table.TransformColumnTypes(#"Time Column Added",{{"ID", Int64.Type}, {"Time", type time}})
in
#"Changed Type" 
Time Dimension in Power Query

Time Dimension in Seconds Grain with DAX:

Run the DAX expression below in a new calculated Table in Power BI or SSAS Tabular model:

Time in DAX = ADDCOLUMNS(
                       GENERATESERIES(1, 86400, 1)
                        , "Time", TIME(0, 0, 0) + [Value]/86400
                        )
Time Dimension in Seconds Level with DAX
Continue reading “Quick Tips: Time Dimension with Granularity of Seconds in Power BI and Analysis Services Tabular Models”

Exporting Data from Power BI Desktop to Excel and CSV – Part 2: Importing Power BI Data Directly to Excel

AAEAAQAAAAAAAAz4AAAAJGQ5ZTk4ZGUxLTI4YjQtNDc4ZC05NTQ4LWRjNDk3OTBlYTE5OQ

Update 2019 April: If you’re interested in exporting the data model from either Power BI Desktop or Power BI Service to CSV or SQL Server check this out. The method explained here is only applicable for Power BI Premium or Embedded capacities with XMLA endpoints connectivity.

In my previous post I explained how to copy and paste data from Power BI Desktop into Excel or CSV. I also explained how easy you can export Power BI Desktop data to CSV using DAX Studio. As I promised, in this post I show you how to import Power BI Desktop data to Excel directly. In this method you don’t need to use any third-party software and the performance is much better than the previous methods.

Note: The method I explain in this post is tested in Excel 2016 only. But, it should work for Excel 2013.

Importing Power BI Desktop Directly to Excel

In one of my previous posts I explained how to connect to a Power BI Desktop from Excel. To import Power BI Desktop data to Excel we have to do the same thing. I explain the way to connect to a Power BI Desktop model directly from Excel, then I show you how to use this method to import Power BI Desktop data.

Finding Power BI Desktop local port number from Power BI Desktop temp directory

We can find Power BI Desktop local port number in number of ways explained here. So in this post I don’t go through all methods.

Whenever we run Power BI Desktop, it opens a random port number. The port number is independent of the model so it doesn’t really matter if  we haven’t connected to any data sources or even if we haven’t open any saved Power BI Desktop (*.PBIX) files. That port number is stored in a text file named “msmdsrv.port.txt”. So the only thing we need is to do is to browse the temp directory of Power BI Desktop and open the “msmdsrv.port.txt” text file. You can find Power BI Desktop temp folder here:

%LocalAppData%\Microsoft\Power BI Desktop\AnalysisServicesWorkspaces

There should be an “AnalysisServicesWorkspaceXXX” folder which XXX is a random number. Open that folder then open “Data” and Find “msmdsrv.port.txt”. Open the file to see Power BI Desktop local port number.

image

Continue reading “Exporting Data from Power BI Desktop to Excel and CSV – Part 2: Importing Power BI Data Directly to Excel”