Jun '23 • General discussion
The product philosophy behind Microsoft Fabric
On May 23rd 2023 (at the Microsoft Build conference), Microsoft Fabric was announced in Public Preview. In this article, we'll dig into the product philosophy behind Fabric to understand the vision of those who created it. But first, let's start with an overview of Microsoft Fabric.
What is Microsoft Fabric?
In the words of Microsoft, Fabric is an end-to-end, unified analytics platforms that brings together all the data and analytics tools that organizations need.
I did an in-depth video about what In the future, I will be making my own video about What exactly Microsoft Fabric IS, but in the mean time, this short video my Microsoft gives a good overview:
The Product Philosophy behind Microsoft Fabric
I found this blog post by Simon Nuss of Hitachi Solutions to be illuminating in terms of the product philosophy adopted by the Microsoft Product Group (the team that have been developing Fabric for the last few years).
Simon and the Hitachi team has been working closely with Microsoft since September 2021 providing testing and feedback on the product throughout its development, so it's interesting to read his insights from 'inside the tent'.
In the blog post, Stephen said:
As one of a handful of experts invited into the conceptual phase, I initially understood Project Trident as just another product codename, like Project Babylon (Purview), or a feature, like Project Athena (Synapse Link). But I quickly discovered it was something else entirely. It was a bold plan by the Product Group (PG) to reimagine the entire Microsoft data and AI stack. The core mission was made clear and has changed little over the course of the past 2-years: SaaS-ify all analytics capabilities into a unified analytics fabric to support low-code plus pro-dev.
This was probably the best description of the mission of Microsoft Fabric.
Let's break it down.
SaaS-ifying Microsoft data and analytics services.
Firstly, Fabric is a SaaS product (software-as-a-service), allowing companies to do pretty much everything they need to do in the analytics space within a nice easy-to-navigate UI in a web browser.
This is a fundamental shift from the Platform-as-a-service (PaaS) business-as-usual offering that Microsoft currently offer. Previously just the Power BI element of the data stack was offered as a software-as-a-service, but on the data engineering/ data science side was only available as a PaaS.
For a business perspective, this shift means a lot less FRICTION on their data and analytics journeys.
Previously, services like Azure Synapse Analytics, Azure Data Lake Storage, Azure Data Factory could only be procured through Microsoft Azure.
This was a barrier for some organizations.
Now, these services have been ported over to Microsoft Fabric and are accessible to business users in much the same way as the Office 365 apps.
In Fabric, you can use these data and analytics resources WITHOUT the need to worry about things like resources groups, virtual networks, storage accounts, Azure AD groups (although for some of the systems in Fabric, you will still need some Azure footprint, like using Azure Event Hubs if you want to take advantage of Fabric Synapse Real-time Analytics).
Low-code plus pro-dev
In the same vein, Microsoft have been Fabric to provide options for both low-code and pro-dev.
They want to make it as easy as possible for people to manage data within their organization.
This a continuation of their (successful) strategy they used throughout the Power BI product. Power BI makes it easy for practically anyone to create dashboards from a variety of data sources, there is a very low barrier to entry.
When you couple an easy-to-use product, with Microsoft's global distribution, adoption rates sky-rocket.
With Fabric, Microsoft have now deployed the same low-code/ easy-to-pick-up philosophy to the data storage and transformation layer of the analytics stack.
That's not to say that the product is basic. It still offers advanced users the opportunity to write advanced SQL views, tables, stored procedures in the Synapse Data Warehouse engine, and also Spark jobs in pySpark, R or Scala.
Knowing what to leave out
A common criticism that I've seen some people make about Fabric in the days after the release is: "isn't Fabric just a re-brand of old technologies?".
To be fair, this was what I initially thought when I first heard about Fabric, but when you spend a bit of time building things in Fabric, it's clear that this is not a fair comment to make.
Yes, existing technologies have been 'SaaS-ified', but it's clear to me that Microsoft has thought deeply about what to port over and how.
For example, if you're familiar with Azure Data Factory, you will soon realise the Fabric Data Factory offers a slim-line alternative (with only a few Activities available). Synapse has also had a major makeover with SQL serverless/ dedicated pools 'simplified' to just Synapse Data Warehouse/ Lakehouse options.
Note: some of these decisions might have been deliberate, and some of them we might see in future iterations of the product (as they are still integrating!).
The AI Revolution
Given the industry context at the moment, with AI becoming mainstream through text generation tools like Chat-GPT by Open AI (in which Microsoft have invested $10b) and image generation tools like Midjourney, it is no surprise to see Microsoft lean heavily into AI when it rolled out Fabric.
Fabric is marketed as 'Data Analytics for the era of AI' - and whilst Fabric offers great potential for AI development with the current set of tools (available in the first public preview release), I feel like there is a lot more to come.
Currently there exists the Synapse Data Science persona/ experience, but it's basically just Spark notebooks, which yes does offer you options to build ML models on top of your lakehouse data, but I wouldn't be surprised to see a lot more development in this space. For example, integrating the the Azure ML Studio tools would fit with the low-code/ visual programming ethos of Fabric.
I would also expect them to lean into their OpenAI partnership to the max and embed Azure OpenAI Service/ Copilot autocompletion type functionality into every code entry part of Fabric: writing DAX measures with Copilot, T-SQL in your data warehouse.
Leveraging Microsoft's global scale
The final point in this look at the product philosophy is all about the rollout. From 01 June 2023, all Power BI workspaces in the globe were given the option to 'turn-on' Fabric.
This distribution advantage is what made Power BI itself become one of the most used BI tools on the market, when it was released to General Availability on 24 July 2015.
Customers like low friction, and Fabric will give them a further reason to remain deeply embedded in the Microsoft ecosystem.
0
0 comments
The product philosophy behind Microsoft Fabric
skool.com/microsoft-fabric-users-7719
A free community where people come together to share tips and tricks to get the most out of Microsoft Fabric.
Powered by