My new Twitter profile.
A couple of months ago I had been reading twitter while eating my lunch at my workplace in Budapest when I saw a tweet informing about Alberto Ferrari going to Prague to hold a training day about data modeling in Power BI, SSAS Tabular and Excel.
I want to thank the organizers of the SQLSaturdayPrague event held at 2018.12.07-08 (with the pre-cons included), they were friendly and attentive. The event was thorougly organized and managed. The sponsors and the fellow attendees deserve a big Thank You (Děkuji vám!) for the event’s friendly, fun atmosphere.
I was a bit nervous before the pre-con. I was afraid of being late so I visited the venue the evening before the event and set-up 3 different alarms to wake me up in time. The event started at about 9 AM and finished about at 5 PM in the beautiful building of the czech savings bank.
It was great to meet Alberto Ferrari in person after reading the books and blogs authored by him and Marco Russo. As Mr. Ferrari said, building a good data model comes first because when you have a well-designed model, you don’t have to write so difficult DAX measures. We talked about different data modeling scenarios and the dangers of using the bi-directional relationship feature of Power BI. I really recommend attending a course taught by Mr. Ferrari, even if you have to travel a bit to get to the venue. Try to finish reading the Definitive Guide to DAX by 2019 february becasue that’s when the 2nd edition is going to be published!
We had three different presentations at the same time so everyone had to choose 7 presentations from the available 21.
Administrating SSRS/PBIRS was a boring task! Sorry… did you said WAS?
The main message was that implementing even a single change is quite slow using the UI of SSIRS/PBIRS. Imagine how much time you can save by doing those using powershell!
Inside the VertiPaq engine
I thought I knew this topic very well because of my relevant MCP exams (70-768) but there were some new elements such as hash encoding and materialization. I am going to research these. If you have resources about a deep dive into the VertiPaq engine, please let me know on LinkedIn.
Monitoring your BI solution with Power BI.
This presentation got a strong applause from the audience because the the solution is easy to implement and very useful. You can use dynamic management views and system views to analyze the logs to check the usage and performance of your reports.
Automating the testing of your BI solutions with NBi. Everyone knows how important testing is. With the open-source NBi
you can create and store tests for different tools such as SSIS and SSAS.
Dbatools – The Swiss army knife for the DBA. Marcos also got a great applause because his presentation was very useful for anyone managing databases. Powershell is awesome, dbatool is a Powershell library if I am not mistaken.
Data Modeling for Beginners
Well, modifying a date model already implemented in a database or DWH is one thing that I have experience in but I never suspected that data modeling is this complicated. Life is complicated, the business processes of banks and other large companies are particulary complicated and databases are trying to store events in a way to be more easily queried so it is no surprise that a lot of data models are more complicated than a starchema.
Predictive Reporting Services Performance Monitoring
The idea is very valueable. Don’t wait for the users to complain. Use R or Python and data science such as clustering to quickly find the problematic reports and their troublesome properties.
After the presentations and the raffle we went out to have dinner with the fellow attendees and organizers.