Tristan Malherbe
@Datatouille
Followers
3K
Following
6K
Media
531
Statuses
4K
β Founder Data Pulse β Co-Founder @ClubPowerBI (Power BI User Group π«π·) & @ClubFabricFR (French Fabric User Group π«π·)
France
Joined June 2013
How I fell in love with #PowerBI and how it has transformed my life ! . A thread π§΅.
8
19
128
The combo Python + SQL is just pure gold for any "Modern" Data Architecture. Python π is so powerful and consumes very few Capacity Units in Fabric. SQL (supposedly dead since 15 years) remains one of the best language to prepare & query data βοΈ.
#Python is eating the world! You can now use Python notebooks to interact with your #SQL database in #MicrosoftFabric. πΒ πΒ πΒ πΒ πΒ πΒ πΒ πΒ πΒ πΒ πΒ πΒ πΒ πΒ πΒ π.
0
2
10
RT @arunulag: I recently had a great conversation with Richard Campbell on @RunAsRadio , episode #993, recorded live at Microsoft Build. Wβ¦.
0
5
0
"The Lost Decade of Small Data" (from @duckdb). π
duckdb.org
We benchmark DuckDB on a 2012 MacBook Pro to decide: did we lose a decade chasing distributed architectures for data analytics?
this chart is basically saying, for a single user use case, a library like #duckdb is competitive with state of the art DWH even at 1 TB scale.
0
1
6
French Riviera, today βοΈποΈ. Perfect environment to think about my future Power BI & Fabric contributions. I have some technical sessions and a table talk scheduled in the next 3 weeks.
2
0
8
Excellent feature here ! . Combining Direct Lake and Import in the same semantic model while keeping "regular" relationships ("strong" and fast) opens up many new possibilities and architectures. I will try this very soon.
The βChoosing Between Import Mode, Direct Lake, and Composite Modelsβ session is available to everyone!.Based on my speech at Fabric Conf 2025 in Las Vegas, I recorded a video using the new "Direct Lake + Import" composite model feature, which is now in public preview. #powerbi
1
2
9
There are now 2 versions of Direct Lake in Fabric: Direct Lake on OneLake vs Direct Lake on Fabric SQL EndPoints (Lakehouse and Warehouse). πThe documentation below gives more info on both and compares them.
We have now updated our MS Learn documentation on Direct Lake to include information on the new Direct Lake on OneLake option that was announced earlier this month at FabCon. #PowerBI #MicrosoftFabric.
0
1
5
Great article from my friend @GilbertQue . Python is usually faster and consumes less Fabric Capacity Units than pySpark when used within Fabric Notebooks. The differences can be huge sometimes. Spark looks like an overkill with low / medium volumes of data, agree ?.
This blog post explains the advantages of using a Python notebook instead of a Spark notebook in #MicrosoftFabric
2
1
7
cc @marcorus.I am sure you will like this one Marco !. Also curious about your thoughts on this @MrSiWhiteley - @PowerBITips π.
2
0
0
Such a great article about current trends in the Data Engineering & Data industry. One of the best I have read in many years. Let's stop this BS please π.Once and for all.
luminousmen.com
Discover the truth behind Data Engineering's buzzwords! Uncover the reality of Data Fabric, Medallion Architecture, Zero ETL, and the Modern Data Stack. Tools don't solve problemsβcraftsmanship does.
1
1
7
Agree with @marcorus !. Microsoft Excel is one of the best software in the world, with an impressive adoption. π Never bet against Excel.π Embrace it.
Honestly, you can't stop people from using Excel, and you shouldn't even try. Excel has been everyone's favorite for decades. Fighting it is a losing battle. Instead, my advice after many years is to embrace it. This is not about Excel. This is about a principle: to help users
0
0
6
RT @marcorus: I usually do not write about announcements and new features until we have had time to try and test them in the real world. Hoβ¦.
0
19
0
RT @arunulag: Tons and tons of exciting announcements at #FabCon check out my blog . #MicrosoftFabric #PowerBI httpβ¦.
0
32
0
"The user is the storyteller, not the report creator" .I love this sentence @kurtbuhler π. Can I steal it from you please ? I will always quote you obviously !.
Be careful with data storytelling in #PowerBI. I think the user is the storyteller, not the report creator. The report is the framework for the narrative. Users choose their own path through it by using interactions. See more:
2
1
12
Never say Never !. I personally use donut charts (which I prefer over pie charts) sometimes. 2-3 categories to be displayed & compared maximum.
π Using pie charts is not the end of the world! . β Why pies/donuts are controversial in #dataviz .β Caveats/use-cases in #PowerBI.β Why focusing on "rules" can be unhelpful. π₯ Ay, ay. put down the pitchfork; check the article/video first! Links below β
1
0
14
RT @patrickdba: Check out the latest updates in the Power BI January 2025 Feature Summary via @MSPowerBI.
powerbi.microsoft.com
Welcome to the January 2025 update! Get ready to elevate your data analysis experience! Weβre thrilled to announce a series of powerful new features designed to make exploring your data easier and...
0
17
0
@cwebb_bi : I know you like the interactions between Power BI & Excel. You should appreciate this extractor π.
1
0
2
I just developped a fully automated extractor using Excel, VBA & Power BI ! . Example below where I get the list of Customers who bought more than X times in a particular year. π. #PowerBI is acting as a database here. And I am sending 'dynamic' DAX queries from Excel.
3
7
54
Changing my mind in the Northern Alps π«π· .ποΈ β·οΈ βοΈ π . 2024 has been the most difficult & most challenging year of my life. I lost my little brother in September and I think about him everyday. Hopefully 2025 will be better. π
Happy holidays to everyone π
1
0
14
RT @MarkPM_MSFT: Excel Mirroring in Fabric / CSV Mirroring in Fabric.In this video I show how I setup Opening Mirroring to build a solutionβ¦.
0
8
0