Our Favourite Picks from Power BI Global Summit 2023

by Danish Ali Detho | O365 & Power Platform Solution Architect///

Power BI Summit is the biggest Power BI conference of the year which ran virtually from 6th March to 10th March. It brought speakers from the Microsoft Power BI team’s product group, community experts, and MVPs from all around the world hosting more than 100 sessions at this conference. The conference has something for all levels from beginners to experts, Power BI Developers to Admins and Citizen developers with sessions covering a large number of topics including Power Query, DAX, Dataflow, Visualization, Power BI Desktop, Power BI report server, Mobile experience, Embedded, Power BI Premium, Architecture, and governance. In this blog, we will cover our favourite sessions from the Power BI Global Summit 2023.

Analyse 130bn + rows in seconds using Azure Synapse and Power BI by Leon Gordon, Tim Watson

We often deal with data sets consisting of billions of records which can take significant time to load and then processing it can be a bit tedious as well. In this session, the presenters took a 132 billion-row dataset of real-world anonymized marketing data and took us through how to build a complete end-to-end analytical storage and data storytelling solution using Azure Synapse and Power BI.

The session will cover everything from designing an ELT process using Synapse Pipelines, building a data model optimized for big data in SQL Dedicated Pool as well as how to manage dual connection mode and aggregations Power BI!

Some of the main highlights include

Harnessing large data volumes using Azure Synapse
How this approach can break down data silos in your company
Combining Import and Direct Query modes within the same Power BI data model
How to create a Power BI dashboard that responds in seconds when working with Big Data in Azure Synapse

Building a robust deployment pipeline in Power BI by Michael Johnson

The flexibility to create and distribute content is undoubtedly one of Power BI’s greatest strengths. However, It does create challenges in terms of control, quality, and reliability of the artefacts deployed to production and that’s where DevOps comes in to play. This session is all about managing Power BI in the same way, using the same tools as we manage other custom-built solutions using the key principles from DevOps. It includes an intro to the tools available in the Microsoft ecosystem, including Azure DevOps, PowerShell, and Power BI Deployment pipelines, to build robust, managed pipelines for high-quality deployments

How do we add Power BI to source control,
Where controlled commits execute a series of tests
Manual validations before new features get promoted to the next stage in our deployment pipeline.

Clustering in Core Power BI and with R and Python by Dejan Sarka

Clustering helps you to find groups of cases, for example, typical groups of customers, and also outliers, cases that do not fit well in any of the clusters which is available in Power BI using scatter plot. In this session, Dejan covered the most popular clustering methods including Hierarchical, K-Means, K-Medoid, and Gaussian Mixture Models clustering. It also covered the algorithms behind the different clustering methods and how to implement advanced clustering in Power BI with help of custom visuals, and R and Python code.

Data Warehousing for EVERYONE! Introduction to Power BI Datamarts by Asanka Padmakumara

The introduction of Power BI Datamarts has opened the doors of Data Warehousing to citizen data analysts/business users. They can create their own data warehouse without depending on the IT team anymore. This session was all about Power BI Datamarts which provide end-to-end data ingestion, preparation, exploration with SQL or using the no-code Visual Editor. It does this by taking the existing benefits and functionalities of Dataflows, Power BI Datasets, and Azure SQL Databases and combining them into an easy-to-use solution.

Incremental Refresh and Hybrid Tables in Power BI by Shabnam Watson

Handling large datasets can be a time-consuming activity and this is where Incremental Refresh comes into play. Incremental Refresh makes it possible for Power BI to handle large datasets by partitioning the data into segments and consequently makes refreshes less resource intensive and more efficient.
This session also covered Hybrid tables, which is simple to configure for a table with Incremental Refresh to enable different storage modes for its different partitions. As the name implies, it provides a hybrid approach with a combination of Import and DirectQuery data storage methods allowing historical data to load into Power BI’s memory (import) for super-fast query performance and to leave the most recent data in the backend (DirectQuery) for near real-time results.

Benefits of Incremental Refresh include:

Fewer refresh cycles for fast-changing data are needed.
Refreshes are faster due to only the most recent data that has changed needing to be refreshed.
Refreshes are more reliable by avoiding Long-running connections to volatile data sources aren’t necessary.
Lesser data to refresh which reduces the overall consumption of memory and other resources in both Power BI.
Large Datasets can grow without the need to fully refresh the entire dataset with each refresh operation.

Power BI Scanner API: full-tenant scan of your metadata by Andrea Martorana Tusa

This was a must-have session for Power BI admin because it introduces us to Scanner APIs. The Scanner API provides all the information an admin needs to monitor your tenant’s status and assess running activities using a single API call. It is a set of 4 asynchronous unified scanning APIs designed to perform a full scan of your tenant’s metadata.
The metadata includes asset names to dataset details, DAX measures, lineage, etc. and it’s incremental which means It is capable of detecting and scanning only those workspaces that have changed since the last time they were scanned.
The session covers how to prepare an end-to-end solution using Power Automate to invoke the Scanner APIs and takes us through each API and the settings required on the Power BI admin portal and in Azure AD.

Row-Level Security patterns in Power BI by Reza Rad

This session covers security in Power BI in depth by showing us how DAX and Power BI security comes together to play an important part in a Power BI project lifecycle: Row Level Security. Several Row Level Security patterns were covered including Patterns such as Static Row Level Security, Dynamic Row Level Security, and many variations of the dynamic RLS to cover scenarios like giving different levels of access to managers and employees to the same set of data.

The session also gives an introduction to defining roles and rules in Power BI Desktop easily using the enhanced row-level security editor. With this editor, you can toggle between using the default drop-down interface and a DAX interface. When you publish to Power BI, you also publish the role definitions.