Fabric vs Databricks Day 2: Features
In this article, we evaluate what Fabric has to offer.
Positive
1) Fabric solves the sprawl problem of Microsoft's analytics stack fairly well.
Azure, like its core competitors (AWS, GCP) has many data services it offers. Fabric appears to do a good job at connecting some of the most commonly used data related services.
Databricks does this very well as well.
2) It leverages no-code/low code manipulation of data manipulations well.
The Power Query-like experience for data manipulation is solid. I can see also how this might appeal to Alteryx-type users, especially those that always tried to push as much of the workload as far upstream as possible!
As of this moment, Databricks doesn't have out there that compete with this, though their Lakeflow Transform has an opportunity to challenge it.
3) Data Factory's has a lot of connectors out of the box!
It will be interesting to see how Lakeflow Ingest executes here. The closest to this that Databricks offers at this moment would be Lakehouse Federation and/or Spark connectors, and even then, Fabric just has a lot more connectors readily available.
4) Really like the concept behind task flows for the purposes of planning on a new project. Perhaps the most innovative feature!
5) Power BI: The anchor product and primary reason why people consider Fabric.
Negative
1) Generally speaking, Fabric brings very little that is new to the table.
Except for a few features and repurposing of some successful mechanisms, this is mostly everything Azure offered 3 years ago. If Azure's services weren't the right ones from a capabilities standpoint 3 years ago, the same likely applies today for Fabric.
In contrast, Databricks does a great job of either innovating OR creating its own flavor of features offered by competitors. For innovation, you can look at how Databricks executes on ML. For how it learns from competitors, you can see how it has developed DBSQL.
2) I am ALL for the UI experience, but this feels like death by a thousand clicks.
3) There is nothing close to the Unity Catalog.
Purview gives you a sense of control, without you actually having control.
It doesn't even appear that you have the full Purview experience either, but just a limited flavor of it.
4) Great for small businesses with less technical chops, but at scale, I am fairly confident managing it efficiently would be very difficult.
While it might handle data that scales well, it won't handle scaling organizational needs well.
Closing Thoughts...
+If you were happy with Azure's products individually, you will be happy with Fabric, especially if you love PBI. It is great for dashboard driven orgs.
+Great low-code/no-code experience.
-The mother of all vendor lock-ins vs Databricks/Snowflake.
-After watching videos from experienced users + navigating it, I can tell that troubleshooting anything would be extremely hard.
Good option for small businesses, would not advise for larger orgs.
Helpful Links
You can find all of the articles in this series at:
Fabrics vs Databricks Conclusion
Also, would love for you to follow me on LinkedIn as well @ in/JosueBogran and Youtube @ JosueBogranChannel.