Data warehouses contain the most important assets for an organization – their data. As companies grow, so do their needs to analyze all data they generate. Organizations are modernizing their data warehouses to reduce costly on-premises datacenter footprints, while using advanced analytics to improve business outcomes. Moving data to the cloud is no small task. Integrating siloed data across business units for a migration comes with many challenges because business logic and other database objects need to be disentangled when that data moves.
One of the more complex migration tasks is translating SQL code. To convert hundreds of thousands of lines of legacy code across database objects requires data teams to either manually rewrite the existing code or hire a vendor to complete the task. Both require thousands of hours of intensive labor and knowledge transfer of business and system logic, which are prone to human error and incur a large cost.
We are pleased to announce Azure Synapse Pathway to help simplify and accelerate migration for both on-premises and cloud data warehouses to Azure Synapse Analytics.
How does it work?
Azure Synapse Pathway connects to the source system and inspects details about your database objects. An assessment report captures further details on the database objects that can be translated into Azure Synapse Analytics. With Azure Synapse Pathway the source database objects are automatically converted and optimized to T-SQL code on Azure Synapse Analytics. This means your existing code, whether a thousand or million lines of code, will be converted by Azure Synapse Pathway.
As a result of these capabilities, the traditional process of manual code conversion can now be automated in a fraction of the time; all while cutting out manual errors and reducing the total cost of the migration.
This preview version of Azure Synapse Pathway supports conversion of database, schema, and tables from IBM Netezza, SQL Server, and Snowflake today. Soon, we’ll provide support for Teradata, followed by Redshift, BigQuery and Hive based Hadoop solutions. We’ll also expand the surface area of conversion to database views, procedures, functions and more to further simplify the automation of analytics workloads into Azure Synapse.