Is Snowflake in an Azure environment a good idea ? (vs Synapse)

Question

I am considering between using Snowflake or Synapse as a Datawarehouse solution. (DWH).

All my data sources are Azure-based (Cosmos, Azure SQL, Blob Storage...). The team to build the DWH is for now only 1 person.

My knowledge of both alternatives sumarized in Pros and Cons.

Snowflake Pros

• Best fit for simple BI solutions
• Cross-database queries
• Auto-scaling and auto-clustering
• Ability to have multiple databases
• Lower learning curve
• Lower administration needed

Snowflake Cons

• Not an azure product.
• No Github integration
• More expensive

Synapse Pros

• Great integration with Synapse: Cosmos Link and Azure SQL link allow an easy pipeline to the DWH

Synapse Cons

• Higher administration effort (caching, partitions, distibutions keys, performance monitoring)
• More complex product. Higher learning curve

Since the human resources for this DWH are very limited, I see Snowflake as a better approach to its "near-zero maintenance" philosphy.
However, having all my Data Sources in Azure, I have the feeling I would do a mistake by choosing an external provider, since Synapse and Azure integration are just perfect.

• What do you think ?
• Does Snowflake really require lower administration efforts ?
• Does the Azure-Synapse integration benefits overcompensate the lower admin efforts from Snowflake ?

I would love to hear your opinions

Answer 1

I do not use Snowflake or Azure Synapse. I know those from conferences, workshops, and learning, with no actual hands-on...

There are two questions in my mind...

• Why not using Snowflake on Azure?
  https://www.snowflake.com/creating-azure-data-warehouse-snowflake/#content

In Azure there is everything well optimized and structured and all out of one hand... even the security aspect and the Role Based Access is implemented, no access/permission handling in two environment...

Did you try both environments as a PoC?