In this episode, Kellyn Gorman talks about the challenges of getting data out of legacy systems (i.e. relational data bases) into Fabric. She explains that whoever hosts the data wins. She talks about often content talks about the golden path or focuses on the marketing content, but it's much rarer to see content that deals with the difficult edge cases.

We talk about how despite being Software as a Service, in order to learn Fabric you will need to learn networking, authentication, and infrastructure more broadly. We talk about how it's impossible to completely get away from hardware and infrastructure.

• Ballmer does an about face on Linux
• Lamp stack
• James Serra on ways to load data into Fabric
• Eugene's benchmarking post, inspired by Jame's post
• AzCopy tool
• Scott Hanselman on abstractions
• Wolf fence algorithm of troubleshooting

Cathrine explains how there isn't a single solution for architecting your data lake with Microsoft Fabric. We walk through all the different moving pieces of getting started with Fabric and lakehouses. Catherine touches on some different ways of implementing medallion in Fabric. She also makes the point Medallion is not the same as Dev / QA / Prod. Lastly, we talk about source control and branching workspaces in Fabric.

Links

• Fabric February
• Medallion architecture (Microsoft)
• Medallion Architecture (Databricks)
• Fabric concepts (capacity, workspaces, licensing)
• Ouroboros

In this Episode, I interview Kristyna Ferris about the different types of data Movement in Microsoft Fabric. Specifically, we talk about gen 2 dataflows, data pipelines, and Spark notebooks. We see how you start simple and work your way up. Kristyna shares the "faucets first" approach at P3 adaptive.

Links

• Yak shaving
• Rapid development
• Datatypes in Fabric Data Warehouse
• VARCAHR(MAX) preview
• Microsoft decision guide
• Semantic Link - Sempy
• Power Bi file import benchmarks
• Semantic Link Labs
• Data on Wheels blog