Štěpán Rešl

Just a quick confession.. I think I am getting old. My previous temper to be loud, creating "shiny" stuff and speaking everywhere, completely vanished. I am thinking more about what I will invest my energy in. If I don't see potential behind, I am just skipping it without a single regret. 😅

270 parquet files with different schemas that all need to be transformed at once. Even just loading them into PySpark requires 17M because it needs to load these files one by one with schema transformation. Because the differences are too vast for Gluten...

48M14S by PySpark
6M07S by Python with #DuckDB

... that is a difference.

In discussions at SQLBits, I have heard many times that you can't connect notebooks to these KeyVaults... So I needed to find the way... Nice... now I need to write an article about it. This is not exactly what you would expect...

In the past few days, I have needed to handle private IP Access to #Azure #KeyVault from Python Notebooks of #MicrosoftFabric ... honestly I wasn't expecting that this would be such a pain. Not just from the perspective of steps, that you need to do, but also in the case of CU(s) consumption...

Yes, this is one of the two endpoints that works for that.

At the table level. We have encountered many issues with simultaneous writing into one table…

Be aware that if you have a lot of writers, you will be locked out. Lakehouses are not behaving fully ACID-ready…

Waiting 10 seconds is not enough at all. First, please trigger the TSQL endpoint sync and then wait… at least a few minutes if you are reading data through that. If you read them directly, then you can almost directly read them after writing them, but…

One more #SVG after the while... I am no longer sharing them, as I can see that many people over-engineer their reports with them. It is an excellent addition to your visual layer, supporting your message, but it should not be the primary carrier of information. But... nah... do whatever you want...

Hurray!! Timeout message with seconds in #MSPowerBI REST API.

Hm... delays? ... naaah just completely failing requests.

I was trying to get into Microsoft Partner Portal, and I have forgotten the order of words that they have for that website... so I have landed there 😂

ONLY 10 DAYS LEFT.

Power Conference 2025 returns to Prague this December. 5 stages, 30+ speakers, real-world stories on Power Platform, Fabric & more. Call for Speakers closes July 31.

Submit now: sessionize.com/power-confer...

#PraguePowerConference

Congratulations Daniel 🤩

Congratulations Erwin

Thanks! To you too

Congratulations 🥳

🎉 6 Years as a Microsoft MVP! #MVPBuzz

Power Conference 2025: Call for Speakers Power Conference is back for its fourth edition, returning to the incredible Cubex Centrum Prague – the venue that hosted nearly 400 attendees last De...

The last 25 days to submit your sessions for #PraguePowerConference #PowerPlatform #MicrosoftFabric

Link: sessionize.com/power-confer...

If you are not doing that and you are using the Lakehouse T-SQL endpoint, you can never trust that the data you are looking at is current.

Yes

Honest question:
Are you using #MicrosoftFabric API for refreshing the T-SQL Endpoint of Lakehouse or #MSPowerBI API?

Do you know that you can directly convert from "import" to "directQuery" mode with the #TMDL editor without needing to recreate the table in Power Query? #MSPowerBI

🤣🤣🤣

Now... I can say that it is alive..

Now it is playable xDDD

@koenv.bsky.social

I have already done Doom xD ...

Pokemon in #MicrosoftFabric!!!