Using Lakehouse Data at scale with Power BI, featuring Direct Lake mode
2025TL; DR
Many companies store data in Delta Parquet files for analytics but struggle with complexity and self-service access. Microsoft Fabric’s Direct Lake mode for Power BI enables high performance and low latency over Lakehouse data. Learn to optimize and analyze at scale!
Session Details
Many companies have invested heavily in building data lakes to store large volumes of structured and unstructured data from various sources into Delta Parquet files. These Delta Parquet files can be used for a wide range of Analytics and Business Intelligence applications. Most of these organizations struggle to derive insights from their investments due to the complexity of accessing and querying the data, and how to let self-service users connect to this data in the lake using Power BI.
With the introduction of Microsoft Fabric, an all-in-one analytics solution for enterprises, we now have a better approach for this. In this session, we will explore how to use Lakehouse data at scale with Power BI, using the new Direct Lake connectivity mode. Power BI Direct Lake combines the best of both worlds from Import and DirectQuery mode, and gives us the option for great performance over data in the lake, without introducing additional latency for dataset refreshes.
We will start by discussing the benefits of the Lakehouse architecture and how it can improve data management and analytics. We will then move on to explore how to connect to Lakehouse data using Power BI by combining both of these architecture components and using each of them to their strengths.
We will also cover best practices for optimizing performance when working with large volumes of data, including using data partitioning and query optimization techniques. We will demonstrate how to use Power BI to analyze Lakehouse data in real-time and how to build reports that provide actionable insights for decision-making.
By the end of the session, attendees will have a solid understanding of how to leverage Lakehouse data at scale with Power BI and how to build powerful analytics solutions that can handle massive amounts of data. Whether you are a data analyst, data scientist, or BI professional, this session will provide you with valuable insights into the world of Lakehouse data and Power BI, featuring the new Direct Lake connectivity mode.
With the introduction of Microsoft Fabric, an all-in-one analytics solution for enterprises, we now have a better approach for this. In this session, we will explore how to use Lakehouse data at scale with Power BI, using the new Direct Lake connectivity mode. Power BI Direct Lake combines the best of both worlds from Import and DirectQuery mode, and gives us the option for great performance over data in the lake, without introducing additional latency for dataset refreshes.
We will start by discussing the benefits of the Lakehouse architecture and how it can improve data management and analytics. We will then move on to explore how to connect to Lakehouse data using Power BI by combining both of these architecture components and using each of them to their strengths.
We will also cover best practices for optimizing performance when working with large volumes of data, including using data partitioning and query optimization techniques. We will demonstrate how to use Power BI to analyze Lakehouse data in real-time and how to build reports that provide actionable insights for decision-making.
By the end of the session, attendees will have a solid understanding of how to leverage Lakehouse data at scale with Power BI and how to build powerful analytics solutions that can handle massive amounts of data. Whether you are a data analyst, data scientist, or BI professional, this session will provide you with valuable insights into the world of Lakehouse data and Power BI, featuring the new Direct Lake connectivity mode.
3 things you'll get out of this session
Understand:
- Direct Lake Mode in Microsoft Fabric
- Limitations that currently exist, and how to deal with them
- Which options exist to make sure to use the full potential
Speakers
Benni De Jagere's other proposed sessions for 2026
A Whirlwind Tour of the Fabric Capacity Metrics App - 2026
Copilot in Fabric - Explained for the Administrator! - 2026
Deep Dive into Microsoft Fabric Capacities: Fundamentals, Concepts, and Advanced Strategies - 2026
Demystifying the on-premises data gateway (for Power BI Semantic Models) - 2026
End-to-end Monitoring for Microsoft Fabric - 2026
Exploring permission boundaries in Microsoft Fabric - 2026
Fabric Capacities, beyond the obvious - 2026
Panel Debate: Real-World Microsoft Fabric Administration - Lessons from the Trenches - 2026
Slide Karaoke - Ultimate Showdown! - 2026
Star Schema ALL the things! But why? - 2026
Understanding Fabric Capacities - 2026
Benni De Jagere's previous sessions
Nose-Dive Narratives: Slide Karaoke 2024
Get ready to wrap up a serious day of learning with a dash of humor, spontaneity, and friendly competition! SQLBits presents "Slide Karaoke" where SQLBits speakers reveal their hidden talents while vying for bragging rights. This session promises to be a one-of-a-kind experience that will leave you in stitches and awe, and the speakers scrambling for their non-existent notes!
Asking questions to get meaningful answers
By making sure we're supplying the other party with the compact information they need, and making sure we're "forcing" them to give a decent answer, there's a lot more information we can receive. By getting the information we need, we'll be able to proceed quicker, and getting to results faster.
Setting yourself up for success when submitting to present
During this session we'll cover how you find those elusive speaking opportunities and how you can make the most of it with just a few simple things to keep in mind. Most importantly, what can you do to help understand organizers or selection committees why exactly your session would be a good fit for them, and why you think it would be a good fit for their audience.
Star Schema ALL the things! But why?
Come along in this journey from source to model to report using a practical mindset, thinking about the design decisions and ramifications along the way. At the core of the session lies the message to think about best practices, with the added step to test, assess, and benchmark what exactly they do for you.
You're a Power BI Admin? Let's get your threads aligned for Fabric!
This session will focus on the life of a Power BI Admin moving over to the extended Fabric realm, and keeping their things in check. We'll look at the familiar topics of capacity and workspace management, and how they align with the new workloads that have been introduced. We'll notice there's a very large overlap, as Fabric has been built on the foundations of Power BI.