#DirectLake
Yesterday at dataMinds.be I asked the audience how many experienced disk defrag.

I was talking about Delta/Parquet files in DirectLake. 
>50% raised their hands.

Next time, I'll ask about index rebuild on SQL Server.

We are so back.
October 8, 2025 at 10:16 AM
Full room for Benni‘s talk on DirectLake mode

#DataCommunityAustria day
January 24, 2025 at 9:36 AM
New Blog: Creating optimal Delta tables using @pola.rs in #MicrosoftFabric Python notebooks.

fabric.guru/delta-lake-t...
Delta Lake Tables For Optimal Direct Lake Performance In Fabric Python Notebook
Learn how to create optimal Delta Lake tables in Fabric Python notebooks for enhanced DirectLake performance using Polars and Spark
fabric.guru
December 7, 2024 at 4:45 PM
Stop the hassle of interrupted connections in #PowerBI when using #DirectLake. @mmoengen.bsky.social guides you through setting up a fixed identity so your published apps can be leveraged without giving direct access to #OneLake.

Watch on YouTube - guyinacu.be/fixedidentity
March 13, 2025 at 3:50 PM
Mirroring isn't a great solution, at least until it provides an easy way to determine which data has changed. It could be a good solution, with some enhancements. Until then, it's suitable for people building DirectLake solutions directly on transactional data structures (bad) or toy sized systems.
September 24, 2026 at 9:56 AM
🚨 New Video Alert! 🚨

Are you giving your app consumers direct access to the Lakehouse when using DirectLake connection? Ah, bad idea for governance! 😬

Instead, check out Fixed Identity for better security and control! 👇

@guyinacube.bsky.social
#GalInTheCube 🩷

www.youtube.com/watch?v=FFsW...
Connection woes using Power BI Direct Lake and an App???
YouTube video by Guy in a Cube
www.youtube.com
March 13, 2025 at 3:19 PM
My own uneasiness about abandonment comes from being really uncertain about the "shoulds" around Fabric. How much should we rely on SQL Endpoint vs making a "perfect" lakehouse for DirectLake? Much of Fabric is v dynamic (scrolling Preview features list) so picking "right" feels harder than it has.
January 16, 2025 at 7:23 PM
September 25, 2024 at 8:14 AM
Curious to learn more about the intricacies of #DirectLake storage mode for #PowerBI semantic models?

Recently, I wrote an article for @towardsdatascience.com that explains the inner workings and potential use cases for Direct Lake in #MicrosoftFabric.

towardsdatascience.com/50-shades-of...
Everything You Need to Know About the New Power BI Storage Mode | Towards Data Science
50 Shades of Direct Lake
towardsdatascience.com
September 11, 2025 at 10:18 AM
Ever had the need to convert a Direct Lake model to plain ol' Import-mode? There's a #TabularEditor script for that...

www.linkedin.com/pulse/conver...

#PowerBI #Fabric #DirectLake
Converting a Direct Lake model to Import
One of my favorite features of Tabular Editor (both 2 and 3) is its C# scripting capability, which allows you to run code, see the impact immediately, and even undo the changes made by the script usin...
www.linkedin.com
August 22, 2024 at 9:00 AM
My dear #PowerBI and #MicrosoftFabric friends, watch out! XMLA write is now supported for Direct Lake storage mode👇👇👇

https://learn.microsoft.com/en-us/power-bi/enterprise/directlake-overview#dataset-write-support-with-xmla-endpoint
August 5, 2023 at 9:25 AM
Have you heard about #DirectLake in #Fabric? The new connectivity mode that promises Import performance with Direct query freshness! Let me introduce you to data temperature to explore what is happening under the covers!

Read it in my new blog!

data-marc.com/2023/09/28/u...
September 28, 2023 at 8:46 PM
Check out my latest blog and video on how to integrate #PowerApps with #MicrosoftFabric Warehouse and enable writebacks for #PowerBI reports using #DirectQuery or #DirectLake models! youtube.com/watch?v=exJW...
shabnamwatson.com/2024/10/26/u...
November 4, 2024 at 3:10 PM
Yesterday I saw a #PowerBI model here that has 8 Dim tables via DirectLake connected to 13 imported Fact tables (all variations of the same data i suspect); and several had a pivoted layout that required *manual* addition of column references to measures on data refresh via a giant SWITCH statement.
April 24, 2026 at 7:38 PM
How will Power BI Embedded with Direct Lake Mode impact developers and ISVs? Hasan Abo Shally from the Microsoft Fabric team introduces this feature, enhancing embedded analytics with real-time data updates and optimized performance. #PowerBI #DirectLake Read their blog: https://blog.fabric.microso…
Introducing Power BI Embedded with Direct Lake Mode (Preview) | Microsoft Fabric Blog | Microsoft Fabric
As we step into 2025, we’re excited to announce the preview of Power BI Embedded with Direct Lake Mode, a new feature designed to enhance how developers and Independent Software Vendors (ISVs) provide...
buff.ly
January 28, 2025 at 6:03 PM
If I don't say it enough, I love working with Michael Kovalsky. We're cooking up the next iteration of the #SemanticModel migration utility for #DirectLake with built-in logic for #DataflowGen2 fast copy eligible query detection. 1/n
November 17, 2024 at 3:29 AM
I don't want to think about how long it has taken me to work out that to insert dynamic values in a #PowerBI textbox using a DirectLake semantic model you have to manually go and turn on the Q&A functionality in the Semantic Model....
It just works in import semantic models!
June 11, 2025 at 1:30 PM
It is good that you can still pull the data into a Power BI Pro workspace, but I am not sure how DirectLake mode is possible in this situation. What am I missing?
January 14, 2025 at 11:10 PM
As far as *Power BI* goes, probably not. Once you start getting above those sizes you are doing some very specific tuning, which Alex here has a lot of personal experience with.

As for Fabric Directlake, it has some specific guardrails you'll quickly bump up on.
learn.microsoft.com/en-us/fabric...
Direct Lake overview - Microsoft Fabric
Learn about Direct Lake storage mode in Microsoft Fabric and when you should use it.
learn.microsoft.com
November 10, 2024 at 12:12 AM