I'm about to develop a new small data warehousing solution using Synapse Analytics. I do not have the opportunity to develop this in Fabric yet. However, I'd like to make sure that I can easily migrate this to Fabric at a later date.
I appreciate that Microsoft have not yet announced what migration tools will be available in the future so I am wondering what will likely be the easiest migration path from Synapse to Fabric?
In the short term, is it best to develop a set of delta-based dim and fact tables using:
I hope that makes sense.
The really quick answer is that I would use Spark for the data ingestion and transformation and then Serverless Pools for the serving layer (pointing Power BI to the Serverless Pool).
Hi @KevinConanMSFT ,
Great thank you. Just so I understand you, do you mean:
- Use Spark (probably Pyspark) to ingest data AND create dim and fact tables
- Use a Serverless db to point to these dim and fact tables to serve these to Power BI (but not do any data transformation to create dim/fact tables within the serverless db itself?)
Hope that makese sense,
If you are looking at that route, I would suggest looking at a combination of Spark Pools and Serverless. That will get you into the mindset of using a Lakehouse which will be easier to port to Fabric. Plus, when you do move it to Fabric, you'll get the SQL Endpoint for the Lakehouse which is an easy way to get T-SQL capability over data managed by Spark!
Great, thanks for that. Would you be able to offer a little more perspective on how you would use Spark Pools and Serverless for designing a data warehousing solution? I.e. what you would use Spark and Serverless components for specifically?
We are building migration tools that will make it easy for you to migrate from a Synapse Dedicated Pool to Fabric.
You can also try out Fabric today to see if it will meet the needs of your new data warehouse.
Great, thanks for the info. However, we don't use dedicated pools.
Are you able to advise which of the other options - e.g. serverless database, lakehouse database or Pyspark scripts would be the easiest solution to develop a data warehousing solution in now and migrate to Synapse in the future?
Check out the November 2023 Fabric update to learn about new features.
130+ sessions, 130+ speakers, Product managers, MVPs, and experts. All about Power BI and Fabric. Attend online or watch the recordings.
Ask your questions about Synapse here!