Europe’s first Microsoft Fabric Community Conference took place in Stockholm, where our Data Domain Lead, Samu, attended to experience the buzz and the latest announcements from Microsoft Fabric. Some time has passed since the conference, but here’s a summary of the key announcements and exciting new features!

For three days, Microsoft unveiled new features and insights at the Microsoft Fabric Community Conference. This post highlights some of the most compelling updates and feature releases across various roles within Microsoft Fabric. But first things first, the conference was huge! Microsoft made an impressive effort, bringing hundreds of team members onsite to share knowledge and address questions across multiple topics.

Key takeaways by categories

The conference offered several tracks and speakers, covering data modeling, governance, analytics, and everything in between. I managed to attend only a fraction of the sessions, but here are some of the most interesting features and updates from my perspective

Platform innovations

⭐ Network security features for all Fabric capacities. Earlier this year Microsoft released the trusted workspace access to access Azure resources like ADLS through the firewall. The downside of this and other security features was that they needed F64 capacity or higher to get them available. So if you had plans to set a smaller capacity into dev and test environments, but still utilizes these features, Fabric says no ⛔ Now Microsoft announced that these features are available for all capacity levels. With this change, all capacity levels can now access these essential security features, streamlining work across multiple levels—a trend we hope continues 📈

New UI experience. It is not once or twice that I have heard the UI is somewhat confusing. To ease up the pain, Microsoft is simplifying at the roles which you can take in the portal. Currently, you can select from six different experiences as it can be shown in the image.

In the future, there is going to be just two different roles; a developer and an analytics one. For me, that makes a lot of sense and I really haven´t find use case for all of these different roles.

Data Warehouse enhancements

For the data warehouse item in Fabric, the main focus were improving the performance by rolling out features like data clustering and result set cache. The other aspect was to bring most of the familiar T-SQL capabilities into Fabric Warehouse like removing the limit for VARCHAR column type or introducing nested CTEs. These and other smaller features are coming according to Microsoft still this year. But there were also couple bigger interesting things that were mentioned.

Database Migration Experience. This service was briefly shown and caught my interest. Part of the Warehouse experience, it allows users to migrate database structures and data from other databases into Fabric. Currently in Private Preview, it’s something I plan to test as soon as possible!

T-SQL Notebook Support. Never wondered could T-SQL be written in Notebooks? Now you can! In Public Preview Notebooks now support also T-SQL language. This enables to centralize all of the code handling in notebooks regardless if it is Python or T-SQL based which helps for example in version control side.

Data Engineering Updates

Fabric Runtime 1.3 Generally Available. Fabric Runtime 1.3 is now generally available as the default for new notebooks. Along with performance and efficiency enhancements, it includes an upgrade to Delta Lake 3.2 and Native Execution Engine support. In our preliminary tests in customer cases, we have seen a significant performance improvement.

High concurrency Mode for Notebooks in Pipelines. Not a groundbreaking new feature, but definitely needed one. This reduces waiting time for on-demand Spark pool spin-up. Basically if you are using custom pool configurations, it needs to start a Spark pool to operate on. This might take some minutes and needs to be done for each of the notebooks. But this feature enables to reuse the session between other notebooks, increasing the overall performance of the pipelines.

Data Science Developments

The biggest hype around data science in Fabric was focused on the AI Skills items. What is AI Skills you ask? In a nutshell, it is a unique Q&A chatbot that takes selected tables from Fabric and the user can ask specific questions to get data-backed answers from those tables. You can enrich the functionality of the AI Skills item for example, by providing pre-defined queries which can be used when a specific question is presented. The demo that Microsoft showed had one of these AI Skills items shared for users in Teams and it did look cool. Have to keep an eye on this one as it does enable interesting use cases. 👀

Data Factory Advancements

Copy Job Task. Although this is not specifically a part of Fabric pipelines, it affects highly especially on the usage of it. Copy Job task was released for public preview in the conference. This is a separate item in Fabric workspaces to ease up the data ingestion into Fabric. This item supports both full load and incremental load jobs, and providing user-friendly experience to set up data ingestion activity. For ad-hoc and testing this can be more faster way to set up the data ingestion part, but have to see what is the performance and scalability for this item.

Invoke remote pipelines. Pipelines in Fabric have been able to call other Fabric pipelines for a while. Microsoft released a highly anticipated feature to call existing pipelines in ADF and Synapse for public preview. This attaches more tightly the existing solution in Fabric by enabling to manage and monitor these different pipelines in a single point.

Real-Time Intelligence

The real-time intelligence side is on 🔥 right now! Microsoft is working hard to bring new features and stability to the streaming experience. Microsoft introduced in the conference many usability improvements, such as the new UI experience in the Real-Time Hub or the UI to manage KQL databases and enabling new connectors for Eventstreams like SQL Server on VM CDC and Apache Kafka.

One of the sessions included a sneak peak into a new query acceleration feature. This enables mixing lakehouse delta tables into a kql database and join data between these two items together. No news when this is coming still, but gives exciting possibilities to couple stream and analytical data faster in the future.

Wrapping Up the #FabConEurope Experience

These are just a few highlights; there were hundreds more, with new announcements released even after the conference. Overall, I was impressed by Microsoft’s commitment to the event, from the venue to the extensive access to Microsoft Fabric experts. I brought back a wealth of information, new connections, and, most importantly, swag!