FabFest is a great community conference held in Birmingham, UK, that Purple Frog Systems has sponsored for the past 2 years. This year, we had the opportunity to host 2 clinics for attendees, where they could learn more about Microsoft Fabric or AI solutions, or troubleshoot issues currently faced in these areas.

Laura is seated at a desk with an open laptop covered in colorful stickers, pointing toward a large monitor displaying a presentation slide. The screen shows the text “ARE YOU FABRIC READY?” in bold letters, along with bullet points: “Make more insights,” “Have confidence,” and “Drive business.” The setting appears to be a modern meeting or training room.

Why did we choose to run these clinics?

At last year’s conference, we had several attendees approach us with questions about Microsoft Fabric:

  • “Where do we start?”
  • “What are the costs?”
  • “How do I solve this issue I’m having?”

By offering a dedicated drop-in room where anyone could sit down with one of our team members, we were able to answer more questions in depth and work through specific technical challenges. It also provided additional activities for people to take part in during breaks or sessions they chose not to attend.

What was the outcome?

Throughout the day, attendees stopped by the room, most often to discuss their existing Fabric solutions rather than to learn the basics; this is likely because the event is aimed at those already using Microsoft Fabric that want to enhance their knowledge. Together, we reviewed their current setups, recommended ways to optimise and troubleshoot issues, and explored how certain aspects of Microsoft Fabric could work better for their organisation.

Some visitors were also interested in learning how to migrate their existing solutions to Fabric and how to make the most of the platform’s capabilities. There were some key questions and topics that were raised by multiple individuals, so I have summarised the advice given.

Key Questions Asked

1. How do I implement a company-wide semantic model and have separate access for different departments?
I recommend utilising multiple workspaces, with one overarching general workspace that stores all company data. Then, create lakehouses in dedicated departmental workspaces that use shortcuts to the main data store, including only the tables relevant to that business department. You can then apply workspace-level and role-based access controls to ensure proper data security and governance.

2. Would you recommend using notebooks for data transformation?
Absolutely! Notebooks are an excellent way to process large volumes of data, access bespoke data sources (including APIs), and are particularly suited to those familiar with Python or PySpark. Having a separate notebook per table helps break up your ETL processes, making them more adaptable and robust. You could even run multiple notebooks simultaneously using high-concurrency sessions to speed up load times. Check our guide on how you can implement notebooks in MS Fabric.

3. How does Git integration work with branching?
When connected to a Git repository, a new branch is made by using the “Branch out to workspace” feature. This workspace becomes your development environment, and you can issue a pull request to merge changes back in. Since branching occurs at the workspace level, switching branches changes it for all users in that workspace, and any uncommitted changes will be lost if you try to switch the branch that a workspace points to. For this reason, it’s best practice to create a separate development workspace when branching and reminding developers that switching branches in a workspaces applies the change to all users.

Final thoughts

Overall, attendees who joined our clinics found them an great use of time, as the sessions were able to be personalised to their specific needs. It was a rewarding experience for us and something we’ll definitely look to continue at future FabFest events!

Tags: , , , ,

« Previous Post
Next Post »