Microsoft has recently taken us on a word lingo with the launch of Azure Synapse Analytics. At Kohera, we’ve experienced some confusion in the market between Azure Synapse as a standalone product and Azure Synapse Analytics as a SaaS platform.

We guess it’s time to set things straight once and for all, or at least for a while.

Azure Synapse Analytics

Azure Synapse Analytics is an integration layer on top of the Azure Synapse SQL Pool. The additional layer is in public preview mode, where the SQL Pools (Azure Synapse) are mature and production ready.

Azure Synapse Analytics offers:

• The mature Azure Synapse SQL Pools
• Perfect integration between various cloud components in one SaaS suite
• SQL serverless (pay-as-you-go data exploration of files on data lake, simple data transformations and logical DWH)
• Perfect integration between the data lake files and Synapse databases
• Synapse Spark which allows you to develop in a Spark environment using Python, PySpark, Scala and C#. Synapse Spark is the MS implementation of the open-source Apache framework. It includes Delta Lake too. Azure Databricks is based on the open-source Spark and Delta Lake frameworks. Azure Databricks has made several improvements to the open-source framework.
• Synapse pipelines (identical to the standalone Azure Data Factory)
• Power BI integration in the Synapse Analytics SaaS suite

Azure Synapse

As mentioned, the core component of Azure Synapse Analytics is Azure Synapse. Azure Synapse is a mature standalone tool. It is a rebranding of Azure SQL Datawarehouse Gen2 and finds its roots in the SQL Parallel Datawarehouse. Azure Synapse is a specialized tool for heavy-duty DWH workloads.

We took Azure Synapse for a spin and we would like to share some lessons that we’ve learned about Synapse in the document below.