On-Premises Data Gateway for Power BI

We have seen different Power BI implementation options in my article here. If you do not want their application or database infrastructure to cloud will be looking for options for generating visualizations from Power BI service(cloud) connecting to on-premises data source (database such as SQL Server, Oracle or even structured files such as spreadsheets or csv files.

Note : Complete on-premises solution is available with the release of Power BI Report Server which is part of Power BI Premium. This article assumes the Power BI service implementation as below and gateway component is highlighted.

Power BI Gateway 1

On-premises data gateway allows Power BI service to connect to your database to refresh dataset. Gateway allows both scheduled refresh and live/direct query.  Time for looking at the steps on how to install and configure the gateway.

Download the standalone data gateway and start the service in on-premises. To download, use the option below after logon to your accountPower BI Gateway Installation_1

After installation, you have to sign in using your Azure account and specify a name to your gateway so that Power BI service tie it to your Power BI account.

Power BI Gateway Installation_3

Power BI Gateway Installation_4

Make sure that On-Premises Data Gateway is running as a windows service in your PC. Ideally this will be one of your on-premises server with connectivity to your database.

Go to Power BI web login, select ‘Manage Gateway’ option as shown below

Power BI Gateway Installation_5

Note that your Data Gateway is already listed there. You need to create a data source and provide authentication details for database connectivity (Windows User details in case of a file data source such as excel spreadsheet)

Power BI Gateway Installation_6

Next, go to your workspace, select the data set and go to ‘settings’ as shown below

Power BI Gateway Installation_7

Under ‘Gateway connection’ section, select your data gateway and click ‘apply’

Power BI Gateway Installation_8

Under ‘Scheduled refresh’ section, turn on ‘Keep your data up to date’ and specify refresh frequency as below

Power BI Gateway Installation_9

That completes it. Your dataset is ready for scheduled refresh connecting to your on-premises data source!

For verifying, you can update data in SQL Server database (local server) and wait for the refresh to happen.

Power BI Gateway Installation_11

Once complete, launch your visualization which uses the dataset and see your updated data reflecting there!

Power BI Gateway Installation_10

Data Analytics – The Power BI way

In my previous articles related to BI, we have explored extensively on Tableau and SAP Business Objects. While I continue the series, it would be unfair if I don’t cover another ’emerging’ solution – Microsoft Power BI (in fact it is a market leading BI solution already, refer to my post for 2017 BI trends here Gartner Magic Quadrant for BI and Analytics)

Those who are familiar with Tableau will find themselves home with Power BI in many ways. However Power BI is unique in terms of features, the publish options, pricing and visualization options such as custom visualizations.

Power BI was completely cloud based solution until recently – i.e, until the release of Power BI Report Server in this month (June 2017).  In the cloud based solution, you have to publish your visualizations to Power BI Service on cloud. You have alternative to Power BI desktop for development in the form of Power BI Service Web Interface which provides similar options for creating and customizing visualizations.

I have sketched a high level cloud based (Power BI Service) implementation architecture below  connecting to on-premises database to understand the different components better.

Power BI Architecture

Implementation architecture using Power BI Service on cloud connecting to on-premises database

The implementation architecture below shows the on-premises solution using Power BI Report Server. In this implementation, development and publishing do not require cloud. Note that the development require installation of custom release version of Power BI Desktop (Power BI Desktop optimized for Power BI Report Server)

 

PowerBI Architecture_OnPremise

On-Premises implementation architecture using Power BI Report Server

In terms of connectivity, Power BI can connect to almost every data source. However the type and location of data should decide your design and type of connectivity.

On-Premises Database connectivity from Power BI Service : If you have an on-premises database, Power BI service allows the connectivity using gateways. The options are (a) data imported to Power BI service or (b) use direct query or live connections. We will explore these implementation models in detail in the upcoming articles.

PowerBI_DataSourcePowerBI_DataSource1

Custom Visuals : This is really great feature that Microsoft lets Power BI developers to create and publish their custom visualizations which can be added to your visualization tray and integrate with your reports/dashboards. Thanks for the support to creativity from developer community.

R Integration to Power BI :  Similar to Tableau R integration, we can connect to R for performing complex predictive analysis.

PowerBI_R_Integration

I will cover R integration details and many more in upcoming posts.