Data Factory in Microsoft Fabric FAQ

Azure Data Factory (ADF) and Azure Synapse pipelines maintain a separate Platform as a Service (PaaS) roadmaps. These two solutions continue to coexist alongside Fabric Data Factory, which serves as the Software as a Service (SaaS) offering. ADF and Synapse pipelines remain fully supported, and there are no plans for deprecation. It's important to highlight that, for any upcoming projects, our suggestion is to initiate them using Fabric Data Factory. Additionally, we have strategies in place to facilitate the transition of ADF and Synapse pipelines to Fabric Data Factory, enabling them to take advantage of new Fabric functionalities. You can learn more about this here.

As we strive to bridge functionality gaps and incorporate the robust data pipeline orchestration and workflow capabilities found in ADF / Azure Synapse pipelines into Fabric Data Factory, we acknowledge that certain features present in ADF / Synapse pipelines might be essential for your needs. While you're encouraged to continue utilizing ADF / Synapse pipelines if these features are necessary, we encourage you to first explore your new data integration possibilities in Fabric. Your feedback on which features are pivotal for your success is invaluable. To facilitate this, we're actively working on introducing a new capability, enabling the migration of your existing data factories from Azure into Fabric workspaces as well.

We don't backport new features from Fabric pipelines into ADF / Synapse pipelines. We maintain two separate roadmaps for Fabric Data Factory and ADF/ Synapse. We evaluate backport requests in response to incoming feedback.

The main function of Fabric pipeline is similar to Azure Synapse pipeline, but by using Fabric pipeline, users can apply all the data analytics capabilities in the Fabric platform. Notable differences and feature mappings between Fabric pipeline and Azure Synapse pipeline can be found here: Differences between Data Factory in Fabric and Azure.

Data Factory helps you solve complex data integration and ETL scenarios with cloud-scale data movement and data transformation services while data engineering helps you create lake house, use Apache Spark to transform and prepare your data. Differences between each of the Fabric terminologies / experiences are available under Microsoft Fabric terminology.

Fabric monthly updates are available at the Microsoft Fabric Blog.

To facilitate customers' transition to Microsoft Fabric from Azure Data Factory (ADF), we offer a range of essential features and support mechanisms. Firstly, we provide comprehensive support for the majority of activities used in ADF within Fabric, along with the addition of new activities tailored for notifications, such as Teams and Outlook functionalities. Customers can access a detailed list of available activities in Data Factory within Fabric. Additionally, we had introduced the Fabric Lakehouse / Warehouse connectors in Azure Data Factory, enabling seamless data integration into Fabric's OneLake environment for ADF customers. We also provide a guide for ADF customers which helps to map your existing mapping data flow transformations to new Dataflow Gen2 transformations. Looking ahead, as a part of our roadmap, we are including the capability to mount ADF resources into Fabric, allowing customers to retain the functionality of their existing ADF pipelines on Azure while exploring Fabric and planning comprehensive upgrade strategies. We are collaborating closely with customers and the community to determine the most effective ways to support the migration of data pipelines from ADF to Fabric. As part of this effort, we will provide an upgrade experience that empowers you to test your current data pipelines in Fabric through the process of mounting and upgrading them.

Microsoft Fabric capacity admins can use Microsoft Fabric Capacity Metrics app, also known as the metrics app, to gain visibility into capacity resources. This app enables admins to see how much CPU utilization, processing time, and memory are utilized by data pipelines, dataflows, and other items in their Fabric capacity-enabled workspaces. Gain visibility into overload causes, peak demand times, resource consumption and more and easily identify the most demanding or most popular items.

The Power Query activity within ADF shares similarities with Dataflow Gen2, but it has extra features that enable actions like writing to specific data destinations etc. This comparison aligns more fairly with Dataflow Gen1 (Power BI dataflows or Power Apps dataflows). Have a look here for more details: Differences between Dataflow Gen1 and Dataflow Gen2.

Using the on-premises Data Gateway, customers can now connect to on-premises data sources using dataflows and data pipelines (preview) with Data Factory in Microsoft Fabric. To learn more, refer How to access on-premises data sources in Data Factory.

Presently, the VNet gateway offers an injective method to seamlessly integrate into your virtual network, providing a robust avenue for using private endpoints to establish secure connections to your data stores. It's important to note that the VNet gateway only accommodates Fabric dataflows at this moment. However, our upcoming initiatives encompass expanding its capabilities to encompass Fabric pipelines.

Fabric Data Factory allows you to develop pipelines that maximize data movement throughput for your environment. These pipelines fully utilize the following resources:

• Network bandwidth between the source and destination data stores
• Source or destination data store input/output operations per second (IOPS) and bandwidth This full utilization means you can estimate the overall throughput by measuring the minimum throughput available with the following resources:
• Source data store
• Destination data store
• Network bandwidth in between the source and destination data stores Meanwhile, we continuously work on innovations to boost the best possible throughput you can achieve. Today, the service can move 1 TB TPC-DI dataset (parquet files) into both Fabric Lakehouse table and Data Warehouse within 5 minutes - moving 1B rows under 1 min; Please note that this performance is only a reference by running the above testing dataset. The actual throughput still depends on the factors listed previously. In addition, you can always multiply your throughput by running multiple copy activities in parallel. For example, using ForEach loop.

You can separate the different workloads between workspaces and use the roles like member and viewer to have a workspace for data engineering that preps data for a workspace that is used for report or AI training. With the viewer role, you can then consume data from the data engineering workspace.

Our current focus involves the active development of CDC capability within Data Factory In Fabric. This forthcoming capability empowers you to move data across multiple data sources combining different copy patterns including bulk/batch copy pattern, incremental/continuous copy pattern (CDC) and real time copy pattern into one 5x5 experience.

In certain user experiences, you may encounter system artifacts not intended for interaction. It's best to disregard these artifacts, as they will eventually be removed from the Get Data experiences in the future.

Data Factory pricing in Microsoft Fabric provides a comprehensive guide to how costs are calculated for Data pipelines and Dataflow Gen2. It includes several pricing examples scenarios to help you understand the pricing model better.

What's new and planned for Data Factory in Microsoft Fabric provides insights into upcoming features and their estimated release timelines over the next few months.

Next steps

Introduction - Data Factory tutorial