It can be quite infuriating to be a Power BI Pro user and constantly be on the verge of doing something, before realising that its only available with Premium.
Thank goodness for Power BI Premium Per User for smaller companies. At just double the price per user licence, all these enterprise level goodies can be available. I wanted to do a whistle stop tour of each one, and some others I have found along the way
Automated Machine Learning
No code machine learning is great if you arent a data scientist and you want to try out some basic machine learning.
For this example I have gone to my dataflow in Power BI Service
I am going to use my date dataflow just as an example, but obviously this isn’t the datasource you would want to use
I wont go any further with this example as its not a great one but I just wanted you to get a feel of the first few screens. This is an incredibly powerful feature. once set up you can use this dataflow in Power BI.
You can also integrate Azure ML models that your data scientists have created in Azure with the Azure ML API
I have previously used Cognitive Services in Azure to add sentiment analysis and key phrases onto tweets as they come in via an Azure Logic App trigger.
This information is then loaded into Power BI Pre prepared.
I have struggled to access this functionality within Power BI previously but hopefully this will allow Power BI users proper access to this fantastic Azure Service
Within Power BI Query Editor (Dataflows in Service)
If you are in Pro you will see the following
With Premium however you can use the following functions
Just a couple of examples, you can take descriptive data and score it with ScoreSentiment. Then you can take these scores and add a new column for Sentiment (Positive, Negative, Neutral)
You can add a language field that will tell you if your test item is for example, English, German, French etc.
And you can extract key phrases from descriptive text to add a cool word cloud to your power BI report.
No code data prep and scale
Visual Data Prep is another area that needs looking at in more detail but this is available in Premium dataflows
You now get a great diagram view of all the actions done on your queries.
This will make things much easier for users who can now interact with the diagram view.
Assign your development App Workspace and Pipelines will then let you create Test and Production App Workspaces. Each workspace can be pointed to different data sources.
You can then very easily work in development. move your work to test and then to production when its ready to use.
As at April 2021 Dataflows haven’t yet been added but once this is in, I think this is really the only way to go with Power BI.
Having done some research on Dev Test and Prod in Power BI Pro this makes a huge difference as its practically unworkable in Pro
Integration with Azure Synapse
Azure Synapse is the new Azure Data Warehouse.
You can now create Power BI reports over big data in Azure Synapse.
And Azure Synapse gives you even bigger wins with performance
Pixel-perfect paginated reports
Report Builder is free. You can download it and even use a model previously created in Power BI.
Here you can build your Pixel Perfect reporting, invoices are a good example. Reports that require page numbers and headers and footers that are most likely to be printed out.
Without Premium you cant publish the rdl files to the Power BI Service.
With Premium, you can publish your pixel perfect report to service along with your other Power BI objects
These Endpoints allow other providers access to your Analysis Services Engine within the Power BI Service. This means that you will be able to use even more tools for governance, monitoring, debugging, modelling etc, rather than only being able to use Power BI Desktop.
Your admin does need to allow this within the Admin Portal
Trillions of rows
With Premium, you can query and analyse 1 trillion rows of data. which is part of the ‘query performance over big data’ feature. In fact any large data set can make use of this feature to make refreshes and response times quicker
This is worthy of a more deep dive feature in future but its certainly exciting.
Here are some of the advanced security features that can be used for your most sensitive data
- Compute isolation
- Double Encryption at Rest
- Double Encryption in transit
- Private links
- Zero egress
- Immutable logs
- TLS 1.2
- Customer Managed Keys
- VNet Network isolation
- Immutable storage logs
- FedRAMP high & DISA IL4
- DISA IL5, DISA IL6 + IDC503
- Vulnerability management
- Automatic Certs rotation
End to End protection
Sensitivity labels and Microsoft Information Protection (MIP)
Sensitivity labels allows you to classify your data across all Microsoft products like power BI, Excel etc.
Its an extra cost and can be purchased standalone or through a Licensing suite like the Microsoft 365 Compliance Plan, Microsoft 365 Business etc.
For pricing you need to contact Microsoft at this time.
If you are using dataflows in Pro, you cant create computed entities such as referencing tables. Merging tables.
This means that dataflows in Pro need to be quite simple. you only get the full functionality in Premium.
This is why most people in Pro prefer to do the data prep in Power BI desktop.
Its becoming more and more difficult for business users to remain with Power BI Pro, all these above features are an absolute must for enterprise level reporting.
It is really worth that switch to either Power BI Premium per user or a Premium node if you are a large enough company