Power BI Insights: Updating measures; Analytical dataflows; ADLS Gen2

April 14 2021

Microsoft Power BI pros share their tips on updating and creating measures without publishing a dataset, leveraging analytical dataflows, and importing data from ADLS Gen2.

Updating and creating measures for Power BI Premium without publishing a dataset

Gilbert Quevalliers, writing on Four Moo, explored how to update or create measures for Power BI Premium users without publishing the entire dataset. He shared an example involving an inaccurate measure. Starting off with an ALM toolkit he deployed granular changes to his PPU dataset. By default, once the system finishes comparing the PBIX file to the dataset it shows all actions.

Quevalliers shifted to using the ALM toolkit for comparison. He wrote:

Once the deployment is complete, I can then see it was successfully deployed. I was then prompted if I want to refresh the comparison. I clicked on No because I know what has been deployed. I could then open my PBIX report (Which is separated from the dataset) and create a report and upload this to the Power BI Service… Finally, I could then create a visual with my new measure…

Using snapshot analysis for Power Apps analytical dataflows

Gil Raviv, writing on DataChant, discussed Power Apps Analytical Dataflows. He recently launched a new app to help users monitor data flows in Power Platform environments, which is connected to ADLS Gen2 Power Apps Analytical Dataflows as a data source. The app issues a snapshot during every refresh spotting any changes.

To install the app you need to have a Power BI Pro account and the permissions from your Power BI administrator to install Power BI template apps from Microsoft AppSource. This app is a Power BI Template App… In the Summary page you can view the main data changes, missing or new records, and find trends of key profiling metrics.

He highlighted preparations, connecting the app, limitations and best practices.

Pulling data from ADLS Gen2 into Power BI

Chris Webb shared how to import data from ADLS Gen2 into Power BI. Users often wonder the best method, debating between Parquet and CSV or Synapse Serverless. For starters, Webb recommends partitioned tables to reach the best refresh performance. However, creating these tables manually can be labor intensive and introduce unwanted errors. Additionally, it can add costs through the use of external tools like Tabular Editor. Incremental refresh offers a free and largely automated way to accomplish this in the background.

Users with a classic data lake may want to consider filtering data before loading, loading it in Parquet files and querying with Synapse Serverless. But for those that don't need to filter data, connecting to ADLS Gen2 promises better performance and reduced costs. Webb explained options like combining data from multiple Parquet files or using hierarchical navigation.

He wrote:

While connecting to tables in Azure Synapse Serverless is reasonably straightforward, connecting direct to files and folders in ADLSgen2 can be quite intimidating for inexperienced Power BI users. As a result exposing ADLSgen2 data stored in Common Data Model format by attaching it as a Dataflow may be worth considering. There’s a performance difference between doing this and connecting direct to multiple CSV files (see here) but it’s certainly a lot more user-friendly. It’s also worth noting that support for the newer CDM manifest format in Dataflows has just been announced, although Dataflows don’t support CDM folders with data stored in Parquet format yet.

FREE Membership Required to View Full Content:

Become a MemberLogin
Joining MSDynamicsWorld.com gives you free, unlimited access to news, analysis, white papers, case studies, product brochures, and more, and it’s all FREE. You’ll also have the option to receive periodic email newsletters with the latest relevant articles and content updates. Learn more about us here
About MSDW Reporter

More about MSDW Reporter