Tag: Power BI

Azure, Azure SQL Data Warehouse, Power BI, Power BI Desktop, Power BI Web App

Azure SQL Data Warehouse and Power BI

Azure SQL Data Warehouse and Power BI

Without a doubt cloud computing is going to change the future of data analytics and data visualisation very significantly. Microsoft Azure SQL Data Warehouse recently released for public preview. Combining Power BI as a powerful data visualisation tool with Azure SQL Data Warehouse will give the users the ability to see data insights of their data stored in Azure Data Warehouse very easily. In this post I explain how to install Azure SQL Data Warehouse and the the way it works with Power BI. Before going any further I’d like to have a look at the Azure SQL Data Warehouse very briefly.

What Is Azure SQL Data Warehouse?

Based on Microsoft documentation a SQL Data Warehouse is

“Azure SQL Data Warehouse is an enterprise-class distributed database capable of processing petabyte volumes of relational and non-relational data.”

Azure SQL Data Warehouse supports stored procedures, user-defined functions, indexes and collations. It uses columnstore index technology which significantly improves query performance as well as getting you up to 5 times compression in compare with traditional row based indexing.

I leave it to you learn more about Azure SQL Data Warehouse. But, it is important to keep in mind that there are some features like primary keys and foreign keys that are NOT supported in Azure SQL Data Warehouse which affect the way we use Power BI as a data visualisation tool over Azure SQL Data Warehouse. Without primary keys and foreign keys there is no physical relationships between the tables so Power BI service cannot detect any relationships by itself. There is a workaround for this that we can create some SQL views in Azure side to make it work. This can be an expensive solution. The other way is to load the data warehouse into a Power BI Desktop model which can detect the relationships automatically.

Now you know a bit bout Azure SQL Data Warehouse let’s get back to the subject and talk more about Power BI and Azure SQL Data Warehouse.

First things first. You need to have a Microsoft Azure subscription. If you don’t already have it you can use it for a one month trial here. You’ll also get $250 credit. But, remember that if you succeed the $250 in less than a month then you’ll need to pay for it if you want to use it longer.

Install Azure SQL Data Warehouse

After you get your Azure subscription, login to your account and you should see a dashboard like this

Install Azure SQL Data Warehouse

I’m not going to explain the above dashboard as it is out of scope of this article.

  • Click New

Install Azure SQL Data Warehouse 01

  • Click “Data + Storage” then click “SQL Data Ware House”

Install Azure SQL Data Warehouse 02

  • Enter a name for your database
  • Select a performance value

Continue reading “Azure SQL Data Warehouse and Power BI” →

On-premises Data Gateway Personal Mode, Power BI, Power BI Desktop, Power BI Web App, Power Query

Power BI and SSAS Multidimensional

Update: If you are here to learn how to browse your SSAS Multidimensional model in Power BI please refer to this article. The current article shows you how to refresh data on Power BI service on top of a SSAS Multidimensional instance based on data import scenario.

It’s been a while that lots of us are waiting for seeing improvements on Power BI and SSAS Multidimensional. The good news is that Microsoft released a new version of Power BI Personal Gateway last week on 3 Sep 2015. One of the new features added to this release is that we can now refresh an on-prem SSAS Multidimensional model (data import scenario) after we published it to Power BI website. But, what data import scenario means? That means we cannot create mashups with data we already have in an existing SSAS Multidimensional database/cube through the SQL Server Analysis Services connector which is available on Power BI website. So we need to connect to a SSAS multidimensional instance through Power BI Desktop and load the cube’s data into the Power BI model. Indeed we will create a relational model on top a multidimensional model from SSAS.

Then we can create reports and publish them to Power BI website and finally we’ll be able to schedule data refresh on the Power BI website.

We can also connect to a SSAS Multidimensional instance through Power Pivot AND/OR Power Query from Excel then load the Excel file into Power BI website.

Note: We can do the same through Power Query, but, we won’t able to setup a data refresh schedule on Power BI website if we didn’t load

It’s just awesome isn’t it?

In this post I show you how to implement all the data import scenarios using Power BI Desktop, Power Pivot and Power Query from Excel.

First of all you need to download the gateway from here. Then you need to uninstall the existing version of Power BI Personal Gateway from your machine and install the new version. The whole gateway installation and process of refreshing an on-prem SSAS database is pretty much the same as what I explained in this post so I leave the installation part to you.  However, I explain the data refresh part again.

SSAS Multidimensional Data Import Scenario Through Power BI Desktop:

Get Data

  • Open Power BI Desktop
  • Click Get Data
  • Select “SQL Server Analysis Services Database” from the list and click “Connect”

Power BI and SSAS Multidimensional 01

  • Enter the SQL Server Analysis Services instance name
  • The database name is optional, but, I put “AdventureWorksDW2012”
  • Click “Select items and get data from Multidimensional or Tabular model”
  • As you can see you can also put your MDX or DAX custom queries, but, we leave it blank in our sample
  • Click OK

Power BI and SSAS Multidimensional 02

Continue reading “Power BI and SSAS Multidimensional” →

Data Warehousing - DW, DAX, Power BI, Power BI Desktop, Power Pivot

Side-by-side Role-Playing Dimensions In Power BI

Role-playing dimension is one those concepts that is discussed a lot from time to time. I also posted an article about implementing role-playing dimensions in Tabular models.

To recap, in the role playing dimensions in SSAS Tabular article I explained three different solutions:

  1. Importing role playing dimensions several times into the model
  2. Creating database views in the source side (in case your source is a from of RDBMS like SQL Server, Oracle etc…) then import the data into the model
  3. Keep the inactive relationships in the model and create several measures to take care of different roles using USERELATIONSHIP functions in DAX

I this post I explain implementation of the third option above. In this scenario you need to create especial calculated measures based on the roles you have in a fact table. One the most common role-playing dimensions is Date dimension. Consider you have to show Internet Sales Amount by Order Date, Due Date and Ship Date in a single chart in your report. In this case, having 3 different date tables won’t help us to achieve the goal.

New to Power BI? Quickly learn about Data Visualisation in Power BI here.

Defining new Measures in Power BI Desktop

Basically, what I’m going to explain in this post is using inactive relationships between FactInternetSales table and the DimDate dimension by adding a new Calculated measure. In this case, we’ll be able to show Sales Amount by different roles, well, dates in this sample in a single chart.

Continue reading “Side-by-side Role-Playing Dimensions In Power BI” →

Power BI, Power BI Desktop, Power BI Web App, Power Query

Power BI and Active Directory for System Administrators

Active Directory and Power BI

One of the most interesting things about Power BI is that it covers a wide range of areas. Therefore, it can help a wide range of different users to analyse and understand their businesses easily. For instance system administrators can use Power BI to analyse  their Microsoft Windows Active Directory. As a matter of fact, Power BI and Active Directory can work together very nicely so that a system administrator can create high level reports and dashboards.

In this , we’ll create a report of the following charts:

  • Total number of computers by Operating System/Service Pack
  • Total number of  computers by year and Operating System
  • Total number of computers
  • Print pages per minute by printer
  • Total number of printers by year and driver name

As a system administrator you can create heaps of other useful reports.

Get Data

  • On Power BI Desktop click “Get Data” then click “More”

Power BI and Active Directory 01

  • Click “Other”, click “Active Directory” then click “Connect”

Power BI and Active Directory 02

  • Enter a Domain name then click OK

Power BI and Active Directory 03

  • As you can see there are 374 tables you can select to create heaps of reports. In this post I use “Computer” and “PrintQueue”

Power BI and Active Directory 04

Continue reading “Power BI and Active Directory for System Administrators” →