Microsoft Fabric is one of my favourite tools to work with.
But if you’re new here, whether as a data analyst or a business owner considering employing Fabric across your organisation, you might be wondering what actually is Fabric and what does it do?
In this newsletter edition, I will break down the fundamentals of Microsoft Fabric and talk about what it is, some of the core components and when to use Fabric.
What is Microsoft Fabric?
Microsoft Fabric is a data analytics platform that handles your entire data workflow. It combines data ingestion, transformation, real-time event routing and visualization through various different workloads. As a SaaS platform, Fabric centralises data storage with OneLake (more on that later!).
Fabric provides a number of integrated capabilities, such as:
- Role-specific workloads – tailored tools for data engineers, business analysts, etc.
- OneLake – a central data store that can be accessed by all Fabric tools
- Copilot – an AI assistant
- Microsoft 365 integration – you can connect Fabric to Excel, Teams and other 365 applications
- Azure AI foundry – provides prebuilt AI models and tools for building custom machine learning solutions
- Unified data management – centralised data discovery that simplifies governance, sharing and use
As a result, Fabric is unified across governance, security and storage, making it ideal for providing insights based on your data.
Core Pillars of Fabric
OneLake
OneLake is at the heart of Fabric.
Think of it like OneDrive for your data – it’s a data lake that is built into the platform and serves as a single storage point. It’s built on Azure Data Lake Storage Gen2.
Some of the benefits of OneLake are the following:
- It removes the need to understand complex infrastructure details like resource groups, RBAC, Azure Resource Manager, redundancy or regions.
- It prevents data silos thanks to serving as a unified data storage system.
- It simplifies data management across an organisation by acting as an overarching umbrella across different users, lakehouses and workspaces.
- It is included in Fabric by default, so no upfront provisioning is needed.
Lakehouses and Warehouses
Fabric supports both lakehouses and warehouses.
The Lakehouse item in Fabric Data Engineering is a data architecture, Spark-based platform, good for engineering, machine learning and large-scale processing. It works for both structured and unstructured data.
The Warehouse item in Fabric Data Warehouse is an enterprise scale data warehouse with open standard format. It features a full SQL engine, is ideal for BI and governed reporting.
To decide which to use, check out the following graphic by Microsoft:
Real-Time Intelligence
Real-Time Intelligence is another one of the workloads offered within Fabric. It analyses data as it arrives (i.e. in real time!) from a variety of different sources and allows this data to be analysed, extracted and visualised.
Fabric gives enterprises real-time visibility without external services through event streams, real-time dashboards and KQL databases.
Data Factory (Pipelines & Dataflows Gen2)
A successor to the Azure Data Factory, Fabric’s Data Factory is now integrated natively, allowing for low-code/no-code transformations, pipeline orchestration and cleaner GitOps and deployment paths.
When it comes to choosing between pipelines and Dataflows, consider the following:
- Dataflow Gen2 is great for flexibility – it’s a low-code interface with over 300 data and AI transformations to help you clean, prep and transform data with ease.
- On the other hand, pipelines are used to create logical groupings of activities that perform a task. This could include calling a dataflow to clean and prep the data. But in general, they’re used when you need rich data orchestration capabilities.
While there is overlap between the two, my advice would be to stick to dataflows for simplicity unless you need a specific functionality found in pipelines.
Semantic Models
Semantic models form the backbone of analytics across Fabric, and have just been updated with the new IQ workload which is currently in preview.
Previously known as Power BI datasets, semantic models simply refer to a way in which we give meaning and context to single, unrelated data. In Fabric, they work with Direct Lake for near real-time reporting and can be shared across teams. If you’re interested in learning how to make a good semantic model, let me know and I can write about this in a future edition of the newsletter!
When To Use Fabric
So, now you know all about what makes up Microsoft Fabric.
But who should use it – and when?
In my opinion, Fabric can be beneficial for practically any organisation that wants to significantly lower operational overhead.
It’s especially helpful for teams that:
-Juggle multiple platforms
-Have to copy data between systems often
-Struggle with fragmented pipelines
-Use real-time data
-Work in cloud-native environments
Having said that, Fabric isn’t an ideal choice for everyone and I’d advise against it if you work with low-latency OLTP systems or strict on-premise-only regulatory environments.
If you need help figuring out how to implement or level up Fabric in your company, feel free to reach out to me.
Latest Fabric News
Even though we’re talking about Fabric Fundamentals, I have to include some of the latest exciting Fabric news from the week:
SAS has made its decision intelligence solution SAS Decision Builder now generally available on Fabric. This tool combines multiple AI models, rules and logic into a single workflow for easy decision-making, allowing users to move from data to action more quickly. It works directly with enterprise data stored in OneLake.
So… that’s a quick guide to the fundamentals of Microsoft Fabric. I can break down any of these topics in more detail in future editions of this newsletter, simply let me know what you’d be interested in learning more about.
Thank you for reading!
About The Author
SharePoint Developer | Power Platform Expert (Power Apps, Power Automate, Power BI) | SPFx | Microsoft 365 | React | 6+ Yrs Exp | GDG Member | Willing to Relocate (US – H1B / UK / Canada) | GDG Gandhinagar Member
