Fabcon Vienna 2025 - Key announcements

-

As the biggest European Fabric conference of the year kicks off in Vienna today we've had a number of key announcements that have been made this morning. Sadly I wasn't able to make this years as I chose to move jobs, but will push for next - it was the highlight of the year last year and have definately missed it this year.

This blog covers all you need to know from this morning, as well as my thoughts.

Graph databases

As the digital twin functionality hinted at, we now get graph databases as a native object in Fabric. The unique power this brings is that data doesn't need to be duplicated into a new object to create these databases - unlike the likes of Neo4j, TigerGraph, etc.

Beyond a couple of specific uses cases (e.g. supply chain analysis, manufacturing lines, tracking data for SAR requests, etc), I've always felt that these are a solution looking for a problem. But now that we have agentic systems, they add a whole load of context that traditional row or columner databases simply can't capture in the same way.

Personally I can't wait to get my hands on them and start to see how much this additional metadata brings to agentic systems.

Geolocation for RTI

This one is a bit more specialist for those that are using RTI. With this feature, you can now use out the box tools to add geolocation visuals over your RTI data and add additional layers and points of interest. Hopefully we got more than the usual US focus, and can get a good number of additional layers (such as population) globally. On top of this, Fabric will have templates for common use cases such as precision farming, territory management, etc - again it'll be good to see global support rather than the typical US focus we've had in the past.

The key with this one is to remember that just because you can now map data, doesn't mean you should. It could well be that something like a bar graph gives you much clearer insight than using a map.

Operational agents for RTI

Think of these as data activator jobs on steroids. They're designed to monitor RTI data flows, find patterns in data, and take proactive action as data comes into Fabric.

They're Copilot based solutions that basically act as a data analysts tailored to specific tasks. the key to getting these working is to provide the right guiderails when setting up the agent - for a while I can see that being a bit of a trial and error process until best practice is established. It sounds really powerful that the agent will find the data and metadata it needs (and the PBI visuals that are relevant), but again getting the guardrails in place is going to be key. And it's not just during the setup that this will need to happen, with reinforced learning included keeping on top of this is going to be necessary.

I'm really pleased to see that the entire thing has been designed to support a human in the loop design with the ability to integrate into workflows (e.g. prompts from the agent appear in Teams). For me that's a really key point as comparitively few decisions should be fully automated - as Toyota originally suggested many years ago with the concept of Jidoka. AI shouldn't replace people, it should augment workflows so that skilled individuals can focus on the higher value parts of their jobs.

Enhancements to Copilot for Power BI

The current Copilot for Power BI experience is being integrated with the wider 365 copilot.

I think this move is great. With the drive to adopt Office 365 copilot, bringing data and insight to users rather than forcing users to change workflows to go to data and insight is the right way to go. My personal experience is that integrating with existing workflows rather than forcing change is a major factor in driving adoption.

Enhancements to Fabric data agents

Alongside the changes to copilot for Power BI, we're seeng enhancements to data agents. The first is that we're getting more support to cover data stored in mirrored databases.

The bigger change is out-the-box MCP suport for Fabric data agents. If you aren't sure what MCP is, have a look at this blog. This is a big step forward that will make it an awful lot easier to integrate structured data into agentic solutions.

Security announcements

We've had a few security announcements made, and I've grouped them together to make it easier to find.

We've heard:
  • Azure private links and Spark outbound access protection are now GA
  • Gateway sypport for database mirroring
  • Preview of workspace level IP filtering
All of which, as with anything security related, are a welcome announcement.

Enhanced Data Warehouse performance

In the past I heard a Microsoft employee refer to the data warehouse element of Fabric as Synapse SQL pools having had the technology equivalent of open heart surgery.

Given how big a change that means, it's great to see that Microsoft are continually working on making this experience better. A 36% performance improvement in 12 months is a great rate of improvement.

Synapse SQL pool migration tool

Speaking of Synapse SQL pools, for those that are currently using it, we now have a native migration tool to make it easier to move.

Personally I'm of the opinion that if you are using Synapse today, now is the time to start looking and planning your Synapse migration. My money is on Microsoft at some point announcing that Synapse will be sunset, and that it is better to move on your terms rather than when we get an announcement as to when end of life will be.

Developer tools in Fabric

We've had a couple of announcements on developer tools today. I'm going to tackle CI/CD first.

To date, I've found the native CI/CD tools available in Fabric to be the weakest part of the platform and really need an overhaul to be enterprise ready. Unfortunately, Microsoft have disapointed in this space again for me. At several of the conferences last year we got told that all Fabric objects would have CI/CD support by the end of 2024. Today we've heard that all items have CI/CD support that is GA. 

As most customers I have worked with will only touch GA features, the original deadline needed to included these features being GA. That means, that for me, Microsoft missed their self imposed deadline by 9 months. I know the original announcement only promissed support and not GA - but for me if you can't use a feature like this in production then it might as well not exist. 

Beyond that we have heard:
  • Preview of the Fabric extensibility framework allowing organisations to build custom Fabric items.
  • User data functions going GA - this is a big one for those that want to use writeback from Power BI and have functions that need to be re-used (no more helper notebooks!)
  • An open source version of the CLI - allowing the community to contribute and expand

And lots more

Beyond this we've had a number of announcements that have been grouped together and hopefully we'll start to see more information as the conference goes on:
  • Enahancements to Fabric databases to cover backup retention, workspace monitoring and Copilot enhancements
  • Preview of custom live compute pools for Spark. Meaning that you can setup always-active spark clusters that can be launched quickly, and still get autoscaling support
  • Preview of item recovery in Fabric. Meaning that we now have a wastepaper bin to protect against that accidental deletion
  • Preview of AI functions in data wrangler. Making it easier for analysts to do things like text summarization, translation, classification, etc
  • Pre-built templates, dashnoards and AI agents for the most common data sources such as SAP, Salesforce, Oracle and D365 to accelerate Fabric adoption
  • Preview of Oracle and BigQuery native mirroring - with SAP to come soon
  • GA of OneLake shortcuts to blob store
  • GA of Onelake catalog Govern tab
  • Preview of OneLake catalog secure tab
  • Preview of the new Copilot experience for OneLake catalog
  • Preview of incremental refresh for materialised lake views - this is going to e a big one for me
  • General availability of enhancements to Copilot in Power BI
  • General availability of Power BI desktop functions in the service - such as editing models, PowerQuery support, composite (DirectLake and import) and creating paginated reports
  • General availability of Purview security integration
  • General availability of surge protection for limiting background jobs
  • Preview of adding workspace level usage limits


Comments

Post a Comment

Popular posts from this blog

Power BI - Fabcon keynote, preview features, and March 2025 announcements

-
Yesterday we had an whole host of new features announced in the keynote ( see my first part in this series for more information). Whilst the main announcements focused on Fabric features, we do have a number of announcements to cover specifically around Power BI.  If this proves popular, I'll be moving to seperating out announcements for Power BI and the wider Fabric platform separately. But before we get going on new features, let's address the elephant in the room ... Power BI Price increases As of today, the list price for Power BI has gone up by a significant amount (e.g. Premium per user going from £15.00 to £18.50 per user per month - about a 19% increase). From what I'm seeing this increase is pretty much causing the pushback that partners warned Microsoft would happen. If you are licensing via O365 E5 licensing you are excluded from this price change. The cloud in the silver lining is that whilst it may not seem like Power BI has had many new features in the last ye...
Read more

Workspace topologies in Microsoft Fabric

-
Image
Since I first heard about Microsoft Fabric during private preview, I have spent many hours pondering the workspace topologies available to us. With this workspace based setup it gives us the power to spin up workspaces quickly and easily. But with that power comes great responsibility. But why is that? As time goes on I can see us struggling from workspace proliferation in exactly the same way that BI has suffered from KPI and report proliferation. That is multiple copies of the same workspace but designed for slightly different purposes. To counter that we need to carefully plan and curate our workspace creation, meaning that we need a set of topologies that we can follow to make our lives easier. But what is a workspace topology? The best way to think of it is a blueprint for how we're going to create workspaces and link them together. As I thought about this more and more, three clear topologies began to emerge. I've called these: Monolith warehouse Hub and Spoke Data Mesh M...
Read more