Python In Microsoft Fabric: Execution Options And Scaling

When the running costs of PySpark is putting you off but you don't know a way around it, Microsoft Fabric is to the rescue. Spark compute isn't essential for all data ingestion and transformation workloads and we can leverage different libraries to optimise performance and costs.

In this session, attendees will see the available options for using python and code-first approaches in Fabric, demonstrated by the pros, the cons and a few use cases. We'll also frame this all into a scalable and forward-thinking pattern that allows cost savings now, and growth in the future.

This session is for those getting started with Fabric who are conscious about cost, performance and using the appropriate tools for the job.

- Present the options for using python within Fabric - Demonstrate performance bottlenecks using different python libraries and compute options. - Provide a scalable framework for adopting code‑first patterns that balance savings today with future growth.