CLOUD REPATRIATION
Bring your data home
Connect to Snowflake, BigQuery, or Redshift as SOURCES. Load transformed data to your on-premises PostgreSQL, MySQL, or SQL Server as the TARGET. Cut cloud compute costs by 60-80% while keeping full ownership of your data pipeline.
REAL-WORLD EXAMPLE
A retail company paying $15K/month in Snowflake compute costs uses DVT to extract analytics-ready data back to an on-prem PostgreSQL cluster. Cloud becomes a source, not the platform.
DATA FLOW
WHAT ENABLES THIS
THE FEDERATION ENGINE →
warehouses are just sources — extraction, compute, and loading run on your hardware
PREDICATE PUSHDOWN →
only the filtered slice you need leaves the warehouse, keeping egress small
SWITCHING THE DEFAULT TARGET →
the on-prem engine becomes your default target with a profile change
RELATED USE CASES
Sound like your situation? You can have the first model running today.