Forum Discussion
Decision on using Fabric SQL or Fabric Warehouse
Hello,
We are trying to migrate from synapse to fabric.
Current solution is:
source data= oracle OCI
use Synapse Pipelines to copy oracle data to ADLS Gen 2.
Synapse Dataflow to transform and destination is Azure SQL db. Serverless. We use this as DWH.
a New client has more than 6TB of data.
What data store we will use as the gold layer where we can build reports:
fabric warehouse or fabric sql db since we are in Azure sql db?
I yet to understand the hyperscale version and if this help chose fabric sql db?
thanks
Hi XhevahirMehalla ,
Thank you for reaching out to fabric community.
- Gold layer: Prefer Fabric Warehouse for your 6+ TB data warehouse and Power BI reporting workloads.
- Bronze/Silver layers: Use Fabric Lakehouse for raw data storage and large-scale transformations.
- Fabric SQL database: Best suited for transactional workloads not as the default choice for an enterprise analytical warehouse.
- Azure SQL Hyperscale: Supports databases up to 128 TB and offers independent compute scaling, but it is an Azure SQL Database service tier not a Fabric SQL database feature.
Recommendation: Use Fabric Warehouse for Gold and validate performance and capacity costs with a proof of concept before migration.
Thanks!!
1 Reply
- v-sathmakuri
Community Support
Hi XhevahirMehalla ,
Thank you for reaching out to fabric community.
- Gold layer: Prefer Fabric Warehouse for your 6+ TB data warehouse and Power BI reporting workloads.
- Bronze/Silver layers: Use Fabric Lakehouse for raw data storage and large-scale transformations.
- Fabric SQL database: Best suited for transactional workloads not as the default choice for an enterprise analytical warehouse.
- Azure SQL Hyperscale: Supports databases up to 128 TB and offers independent compute scaling, but it is an Azure SQL Database service tier not a Fabric SQL database feature.
Recommendation: Use Fabric Warehouse for Gold and validate performance and capacity costs with a proof of concept before migration.
Thanks!!