Last week was Microsoft’s Ignite conference hosted in Seattle and we were on the edge of our seats for some highly anticipated announcements. We’ve been playing and helping shape Fabric through the preview and into the Public Preview announced at Build earlier in the year so it’s great to see Fabric is now generally available. Johnny and I caught up to discuss our favourite pieces of Fabric news from the event and share them with you here. You can find the full run down of announcements in the Book of News, or in Simon’s roundup. For now, Johnny and I are giving you a round-up of Fabric-related news from Ignite.
With the huge news of Microsoft Fabric being announced as generally available it begs the question of what that actually means for end users and there’s a couple of really important things to know:
Of the 7 experiences in Fabric, all but Data Activator (just announced in October) are now out of preview. That doesn’t mean they are finished products though and there’s still a lot of functionality to come, to make this a production ready platform like full source control across all artifacts, on-premises connectivity, and a mature security model in OneSecurity.
GA is the green light for many organisations to start trying out Fabric properly. When in preview, Microsoft’s support model is much reduced due to the lesser stability of the product in this phase. This is often a strict blocker for many organisations even building proof of concepts.
GA also gives us a bit more information on pricing. The pay-as-you-go pricing model was already available throughout the public preview, but the reserved pricing has now also been released. Whilst pay-as-you-go allows you to stop and start your Fabric capacity and only pay for what you use, Reserved Pricing assumes you need your cluster to be “always on”, but by paying in advance to have this available Microsoft will offer a whopping 41% discount. You can check out full pricing details here: Microsoft Fabric - Pricing | Microsoft Azure.
AI as a whole was a hot topic throughout the conference, with news around a new range of next gen processing hardware designed to support AI, as well Microsoft’s commitment to offsetting the carbon footprint of the AI age with large investments in renewable and sustainable energy sources.
One theme you could not escape was Copilot. Copilot is an AI assistant. The data viz enthusiast with me can already imagine what an Ignite word cloud might look like, as it would be pretty much “Copilot” and then many minute and illegible other Ignite related words. Copilot is going to be everywhere in the Microsoft product eco-system and that of course means that it’s also coming to Fabric.
The announcement from Ignite was that this is a preview feature and it’s also worth bearing in mind that it won’t be available by default. Firstly, the rollout will be phased, so you it’s not expected to appear in all tenants until the end of March 2024. Then the preview will be available to those with F64 or P1 skus and above. It still then also needs to be enabled in tenant settings.
Once available expect to see Copilot popping up in Data Science, Data Engineering, Data Factory and Power BI. Here it will offer a ton of productivity assistance where it will write and suggest code for you and even build reports from scratch. Of course this won’t put you out of a job – outputs still need to be vetted and may need some tweaking and refinement, but overall should help data professionals to get things done quicker.
So, this is a really exciting one for me. Distilling it down to a single sentence, Mirroring gives us direct access and replication of database sources into the Microsoft Fabric Data Warehousing experience, meaning you’ve got up to date data from that source database, inside Fabric. I have a wealth of questions and ideas about how this works that we’ll be exploring in an upcoming episode of Advancing Fabric, like – does this change ETL patterns and architecture if we have up to date data readily available, already in Delta Lake format in Fabric. How will this impact the source performance, what source types does this actually work with…and so on. Definitely an interesting feature to keep an eye on.
Microsoft had already announced that they will be holding a new data conference, and based on feedback had also already taken time to switch the dates from December ’23 to March ’24. The announcement at Ignite revealed a re-branding and this event will now be known as the Microsoft Fabric Community Conference.
This re-iterates the fact that Microsoft really do want Fabric to be front and centre of their data offering.
The agenda has yet to be announce yet, but there are some speakers confirmed, with a whole host of familiar faces from the Microsoft team currently slated to deliver talks on everyone’s new favourite data platform. Maybe we’ll see you there?
These are but a few headlines from Ignite. Johnny and I dive deep into the full list of November updates and what they mean in our November News video on YouTube, coming soon.