USE CASES

Real cross-engine problems, solved.

DVT is the fastest path from raw operational data to business insights. Eight ways teams use it — each with the exact capabilities that make it possible.

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.

Snowflake (source) → DVT + DuckDB → PostgreSQL on-prem (target)
Full story + what enables it →

HYBRID ANALYTICS

Best of both worlds

Keep operational data where it lives — on-prem MySQL, Oracle, SQL Server. Join it with cloud warehouse data from Snowflake or Databricks. DuckDB federates the query locally. No data leaves your network unless you want it to.

Oracle on-prem + Snowflake cloud → DuckDB federation → Target of choice
Full story + what enables it →

REPLACE ENTERPRISE ETL

Own your pipeline

Eliminate Informatica, Talend, SSIS, or DataStage licenses. DVT does the same job with SQL you already know — no GUI, no per-connector fees, no vendor lock-in. One tool replaces a $50K-500K/year license.

Any source → DVT (SQL models) → Any target
Full story + what enables it →

MULTI-CLOUD CONSOLIDATION

End vendor lock-in

Sources scattered across BigQuery, Snowflake, and Redshift? DVT federates across all three into one unified target. Switch cloud providers without rewriting your pipeline. Your SQL models are engine-agnostic.

BigQuery + Snowflake + Redshift → DVT → Unified target
Full story + what enables it →

STARTUP DATA STACK

Full ELT on a laptop

Start with zero cloud costs. PostgreSQL + DuckDB + DVT gives you a complete ELT stack on your laptop. Build your data models, test locally, and scale to cloud when revenue justifies it. No $500/month Fivetran bills on day one.

CSV/API → DVT seed → PostgreSQL → DVT models → Dashboard
Full story + what enables it →

ANALYTICS ENGINEERING

dbt + any source

Already using dbt? DVT extends your project to reference sources beyond your warehouse. Add MySQL operational data, Oracle ERP data, or S3 files — without changing your existing models. Same ref(), same source(), same everything.

Existing dbt project → Add DVT → Reference any source
Full story + what enables it →

DATA LAKE OFFLOAD

Build a lake without Spark

Materialize models to S3, GCS, or Azure as Parquet or CSV. Build a data lake from any source without Spark infrastructure, EMR clusters, or Databricks workspaces. DVT + Sling handles the heavy lifting.

Any database → DVT → S3/GCS as Parquet
Full story + what enables it →

ON-PREM BI ACCELERATION

Dashboards in minutes

Connect to ANY operational database — ERP, CRM, POS, MES. Transform with SQL. Materialize insights tables optimized for Tableau, Power BI, or Metabase. Incremental updates mean dashboards refresh in minutes, not hours.

Oracle + MySQL + SQL Server → DVT → PostgreSQL → Power BI
Full story + what enables it →